DOCTORAL THESIS IN MACHINE DESIGN STOCKHOLM, SWEDEN 2015

Innovation under pressure Reclaiming the micro-level exploration space

Katarina Lund Stetler

Doctoral thesis no. 2, 2015 KTH Royal Institute of Technology School of Industrial Engineering and Management Department of Machine Design Division of Integrated Product Development SE-100 44 Stockholm

Innovation under pressure Reclaiming the micro-level exploration space

©Katarina Stetler, 2015

TRITA MMK 2015:02 ISSN 1400-1179 ISRN/KTH/MMK/R-15/02-SE ISBN 978-91-7595-449-3

Printed by: US-AB, Stockholm, Sweden

Academic thesis, which with the approval of Kungliga Tekniska Högskolan, will be presented for public review in fulfilment of the requirements for a Doctorate of Engineering in Machine Design. The public review will be held at Kungliga Tekniska Högskolan, Kollegiesalen, Brinellvägen 8, at 13.15 on the 20th of March 2015.

My passion has been to build an enduring company where people were motivated to make great products. Everything else was secondary. Sure, it was great to make a profit, because that was what allowed you to make great products. But the products, not the profit, were the motivation.

- Steve Jobs

ii

Abstract Research & Development (R&D) departments are becoming increasingly structured and routine-based, with tight schedules and daily follow-ups. This way of working stems from increased demands for delivering products to customers quickly and with high quality at a low price. At the same time, these organisations are faced with the challenge of coming up with new ideas that can become the foundations of tomorrow’s innovations. This means that R&D departments must achieve both exploration, in terms of coming up with new ideas, and exploitation in terms of turning existing ideas into products available on the market. If these dual perspectives are to be met within a single work unit, the employees in that unit must achieve what we call contextual ambidexterity. Previous research has shown this to be difficult to achieve and has offered little guidance for organisations about how to organise and manage their operations in order to increase their chances of achieving contextual ambidexterity.

The aim of this thesis is to explore challenges related to innovation that are encountered at the micro-level in contextually ambidextrous organisations and to shed light on factors that explain those challenges.

This study has combined survey data with interview data from several organisations to analyse the relationship between aspects of efficiency and aspects of creativity. It was found that employees in a contextually ambidextrous organisation struggle to ensure enough micro-level exploration space, in other words, they have trouble finding time to explore ideas and making room for novel ideas.

This research shows that a contextually ambidextrous approach in R&D will likely exert two main challenges related to innovation. The first challenge is a crowding out of exploratory activities in favour of exploitatory activities. One reason for this is the combination of using productivity goals for exploitation and not using any similar targets for exploration activities. Large discrepancies in how these two types of activities are treated runs the risk that the one that is less monitored – most often exploration – is likely to be crowded out in favour of the one that is more intensely monitored.

A second possible challenge is the demand on predictability in project progress that is often built into organisations as a means to enhance exploitation. This aim for predictability might create a reluctance to introduce new projects with high levels of novelty because the introduction of novel ideas contains uncertainties that jeopardise the adherence to the project plan. The combination of this view of novelty in the later phases of product development and the crowding out of exploratory activities could possibly lead to insufficient room for novel ideas to gain ground in the organisation, and this could lead to less innovative output.

Keywords: Ambidexterity, exploration, exploitation, creativity, innovation, micro-level exploration space, operational level, operations, operational effectiveness, employee, lean, lean thinking, process management, streamlining, flow.

iii

Sammanfattning Arbete inom Forskning och Utveckling (FoU) har blivit mer och mer strukturerat och rutinbaserat, med hårt pressade tidplaner och dagliga uppföljningar. Det här sättet att arbeta är en effekt av ökade krav på att leverera produkter till kunderna snabbt och med hög kvalitet till en låg kostnad. Samtidigt ställs de här organisationerna inför utmaningen att generera de idéer som ska bli grunden för framtidens innovationer. Det betyder att FoU-avdelningar måste hantera både utforskande i form av att komma på nya idéer och exekvering i form av att omvandla existerande idéer till produkter och tjänster som kunderna kan köpa. Om dessa två perspektiv ska uppnås inom en och samma avdelning så måste de anställda på den avdelningen uppnå vad vi kallar kontextuell tvehänthet. Detta är, enligt tidigare forskning, en betydande utmaning och det finns få exempel på hur organisationer bör organisera sig och leda sina verksamheter för att förbättra sina chanser att uppnå kontextuell tvehänthet.

Syftet med den här avhandlingen är att utforska utmaningar relaterade till innovation som man möter på micro-nivån i kontextuellt tvehänta organisationer och att belysa faktorer som kan förklara dessa utmaningar.

Genom att kombinera enkätdata med intervjuer i flera organisationer så har förhållandet mellan effektivitetsaspekter och kreativitetsaspekter analyserats. Resultaten visar att anställda i en kontextuellt tvehänt organisation kämpar med att säkra tillräckligt individuellt kreativt utrymme, med andra ord att hitta tid för att utforska idéer och skapa utrymme för innovativa idéer.

Den utförda forskningen visar att ett angreppssätt i FoU som bygger på kontextuell tvehänthet troligen kommer att medföra två utmaningar kopplat till innovation. Den första utmaningen handlar om en utträngning av utforskande aktiviteter till förmån för exekverande aktiviteter. En orsak bakom detta är kombinationen av att använda effektivitetsmål för exekvering medan man inte använder explicita mål för utforskande aktiviteter. Om det är för stor skillnad mellan hur de två typerna av aktiviteter behandlas så riskerar den aktivitet som mäts och följs upp i mindre omfattning, oftast utforskande, att trängas ut till förmån för den som noggrant mäts och följs upp.

En annan möjlig utmaning är kraven på förutsägbarhet i projektframskridande som ofta byggs in i organisationer som ett sätt att förbättra exekveringen. Den här strävan efter förutsägbarhet riskerar att skapa en motvilja till att introducera projekt med hög nivå av nyskapande eftersom nyskapande idéer ofta innehåller osäkerheter som äventyrar möjligheten att hålla sig till projektplanen. Att kombinera det här synsättet på nyskapande i de senare faserna av produktutveckling med utträngning av utforskande aktiviteter riskerar att leda till svårigheter för innovativa idéer att få fäste i organisationen och kan följaktligen också leda till mindre innovation.

ii

Acknowledgements This thesis was made possible thanks to the help and support from a large number of people. First of all, I would like to thank all of those who were involved in making this industrial PhD position at Scania possible. Thank you for giving me the opportunity to immerse myself in an area of research that interests me deeply.

My research journey first took off with Margareta Norell Bergendahl as my supervisor. Maggan, you are a truly inspiring person and your ability to navigate academia and industry is impressive. I am especially grateful for all that you have taught me about how to start up and run fruitful collaborations. As Margareta moved on to new challenges at KTH, Mats Magnusson took over as my supervisor. With Mats’ profound knowledge in the field of innovation management, including the most obscure streams of literature, my research found a stable home in the field of ambidexterity. Mats, besides being a skilled researcher and supervisor you have made conferences and office work more fun with your loud and contagious laughter echoing in the corridors. Thank you!

Other people at KTH also deserve special thanks. First of all, I would like to thank my secondary supervisor, Jens Hemphälä, for good advice along the journey. I would also like to thank Susanne Nilsson and Carl Wadell for being such great office mates. It has been great to share the ups and downs of being a PhD student with you. Thank you also to all the other members of the Integrated Product Development team, past and present. I will miss you!

My other office, at Scania, is also filled with people who have been involved in my research in different ways. Thanks to Ragnar Glav, and earlier in the project Johan Tingström, my industrial supervisors, for helping me balance the different worlds of academia and industry. My managers at Scania, rather many in numbers over the years, have all supported my research and included me as part of the team while at the same time giving me the flexibility that the odd nature of a PhD project requires. Special thanks to my colleagues at RTTI, and all the interviewees and respondents who generously shared their views of innovation under pressure. The steering committee at Scania, headed by Lars Dahlén, has been another important source of inspiration. Learning about the committee members’ views on product development, Scania as

iii

an organisation, and innovation at large has substantially improved the quality of my research.

Over the years I have had the pleasure to co-author publications together with a number of skilful academic professionals. Anna Karlsson, who is a wonder of structure and good ideas, and Jennie Björk, who brought her love for statistics to the table, deserve special thanks. Another co-author is also worth mentioning, although our writing has happened outside of work. Special thanks to Magnus MackAldener, who always sets the bar a little higher. What started out as an idea about a short publication in the popular press ended up as an entire book fully packed with our ideas about creativity in a world of increasing pressure. It has been a great pleasure to work with you!

Many thanks to PIEp and ProViking for providing an excellent platform to meet and learn from other PhD students. Thanks to all the discussants at seminars and anonymous reviewers who have helped advance my research.

I would also like to thank my friends for all the wonderful time spent together at parties, weddings, trips, dinners, and just hanging out. While this PhD project was going on at work, life was spent with you outside of work. Special thanks to my family, both core and extended, for being there, for giving support, and for being who you are.

And finally, thank you Gustaf. With your intense, never-failing, high-energy, witty, smart, supporting, and optimistic way of being you raise me to the skies and make me believe I have earned to be there. With you by my side nothing feels impossible. Du och jag med världen!

Katarina Stetler

Stockholm, February 2015

iv

List of appended papers

A. Karlsson, A. & Lund Stetler, K. (Forthcoming). Frequency versus effects – Obstacles to innovation and their relationship to innovation self-efficacy. Accepted for publication in the International Journal of Innovation and Technology Management. (A previous version was published in the proceedings of the 14th International CINet conference, 8-11 September, 2013, Nijmegen, the Netherlands.)

B. Lund Stetler, K. & Magnusson, M. (2015). Exploring the tension between clarity and ambiguity in goal setting for innovation. Accepted for publication in the Creativity and Innovation Management journal.

C. Lund Stetler, K. (Forthcoming). Creativity just in time? The role of delivery precision in product development. Accepted for publication in the International Journal of Innovation and Technology Management (A previous version was published at the 13th annual CINet conference, 16-18 September, 2012, Rome, Italy.)

D. Lund, K., Björk, J. & Magnusson, M. (2014). Myopic Creative Climate – The Result of Streamlining in R&D Organizations? Published at the 74th Annual Meeting of the Academy of Management, 1-5 Aug, 2014, Philadelphia, PA, U.S.

E. Helander, M., Bergqvist, R., Lund Stetler, K., & Magnusson, M. (Forthcoming), Applying Lean in product development – Enabler or inhibitor of creativity?, Accepted for publication in the International Journal of Technology Management. (A previous version was published in the proceedings of the 14th annual CINet conference, 8-11 September, 2013, Nijmegen, the Netherlands.)

F. Lund, K. & Glav, R. (2014) Strategies for managing micro-level contextual ambidexterity –Combining exploration and exploitation in R&D, Published in the proceedings of the 15th annual CINet conference, 7-9 September, 2014, Budapest, Hungary.

v

List of additional publications In addition to the appended papers, Katarina Lund Stetler has been the author or co-author of the publications listed below. The publications appear in chronological order.

Alfredsson, L., Fazl, A. Lund, K., Söderberg, B. (2011). Product Development Management. Chapter in ”Entering the tigers cave – Perspectives on Japanese and Swedish Product Development, Editor: Dag Bergsjö, Printed in Sweden by Chalmers reproservice, Göteborg 2011. ISSN 1652-9243, Report no: 53

Lund, K., Tingström, J. (2011) Facilitating creative problem solving workshops: Empirical observations at a Swedish automotive company. Proceedings of the International Conference on Engineering Design, August 15-18, Copenhagen, Denmark.

Kihlander, I., Nilsson, S., Lund, K., Ritzén, S., Norell Bergendahl, M. (2011), Planning industrial PhD projects: Speaking both ‘academia’ and ‘practitionese’, Proceedings of the International Conference on Engineering Design, August 15-18, Copenhagen, Denmark.

Lund, K., Magnusson, M. (2011). Slack – a driver of innovation in R&D?, Proceedings of the 12th International CINet conference, September 11-13, Aarhus, Denmark

Karlsson, A., Lund, K. (2012), Knowledge sharing in product development – Exploring the effects of power struggle and task conflict, Proceedings of the International Design Conference – Design, May 21-24, Dubrovnik, Croatia.

Lund, K., Magnusson, M. (2012). The delicate coexistence of standardized work routines and innovation, Proceedings of the 19th International Product Development Management Conference IPDMC, June 18-19, Manchester, U.K.

Hemphälä, J., Lund, K. (2012), Creative climate and continuous improvement – An empirical test of the dimensions in the CCQ, Proceedings of the 13th International CINet conference, September 17-18, Rome, Italy.

vi

Lund, K. (2012) Process management in R&D – Doom or salvation for creativity?, Licentiate thesis, KTH Royal Institute of Technology, Stockholm.

Karlsson, A., Lund, K., (2013), Are you starving your company’s intrapreneurs? Chapter in Innovation and entrepreneurship – A study of innovative clusters in California and Sweden, Editors: Dag Bergsjö & Jennie Björk, Co-editors: Katarina Lund, Daniel Corin Stig, Anders Forslund. ISBN: 978-91-980300-6-8, 2013 (ed.). Printed in Sweden by US-AB, Stockholm, 2013

Table of contents 1 INTRODUCTION ........................................................................................... 1

1.1 CHALLENGES IN THE FIELD OF ORGANISATIONAL AMBIDEXTERITY ............................................ 1 1.2 AIM OF THESIS AND RESEARCH SETTING ......................................................................................... 2 1.3 THEORETICAL POSITIONING .............................................................................................................. 3 1.4 STRUCTURE OF THE THESIS ............................................................................................................... 4

2 ACHIEVING ORGANISATIONAL AMBIDEXTERITY ............................ 5 2.1 DUAL PERSPECTIVES IN R&D ........................................................................................................... 5 2.2 THE AMBIDEXTROUS ORGANISATION ............................................................................................... 6 2.3 EXPLORATION ACTIVITIES – A KEY INGREDIENT OF R&D WORK ............................................... 7 2.4 ACHIEVING EFFICIENT EXPLOITATION .......................................................................................... 11 2.5 COMBINING EXPLORATION AND EXPLOITATION .......................................................................... 16 2.6 THE PARADOXICAL RELATIONSHIP BETWEEN EXPLORATION AND EXPLOITATION ............... 22 2.7 THE CHALLENGES AND BENEFITS OF CONTEXTUAL AMBIDEXTERITY ...................................... 25 2.8 RESEARCH QUESTIONS .................................................................................................................... 28

3 RESEARCH APPROACH AND METHODOLOGY ................................. 29 3.1 OVERALL RESEARCH DESIGN .......................................................................................................... 29 3.2 STUDY 1: SURVEY STUDY ................................................................................................................ 34 3.3 STUDY 2: INTERVIEW STUDY ON THE IMPLEMENTATION OF LEAN IN R&D .......................... 38 3.4 STUDY 3: PRIORITISATION BETWEEN EXPLORATION AND EXPLOITATION ............................. 40 3.5 METHODOLOGICAL QUALITY ASSESSMENT .................................................................................. 41 3.6 REFLECTIONS ON THE RESEARCH PROCESS .................................................................................. 46

4 SUMMARY OF APPENDED PAPERS ...................................................... 49 4.1 PAPER A ............................................................................................................................................ 51 4.2 PAPER B............................................................................................................................................. 52 4.3 PAPER C ............................................................................................................................................. 53 4.4 PAPER D ............................................................................................................................................ 54 4.5 PAPER E ............................................................................................................................................. 55 4.6 PAPER F ............................................................................................................................................. 56

5 ANALYSIS ..................................................................................................... 59 5.1 CHALLENGES ENCOUNTERED IN CONTEXTUALLY AMBIDEXTROUS ORGANISATIONS ............. 59 5.2 I DON’T HAVE TIME! A KEY OBSTACLE TO INNOVATION? ........................................................... 59 5.3 DARE TO WIN – THE ROLE OF RISK-TAKING IN INNOVATION.................................................... 68 5.4 REVISITING THE RESEARCH QUESTIONS ....................................................................................... 71

6 IMPLICATIONS FOR THEORY ................................................................ 75 6.1 INNOVATION UNDER PRESSURE ..................................................................................................... 75 6.2 EXPLOITATORY ACTIVITIES AND EXPLORATORY OUTPUT .......................................................... 80

7 IMPLICATIONS FOR PRACTICE ............................................................. 83 7.1 REASONS BEHIND THE CROWDING OUT OF MICRO-LEVEL EXPLORATION SPACE ................... 83 7.2 RECLAIMING MICRO-LEVEL EXPLORATION SPACE ...................................................................... 84

8 CONCLUSIONS ............................................................................................. 89 8.1 SUGGESTIONS FOR FUTURE RESEARCH ......................................................................................... 90

9 REFERENCES ............................................................................................... 93

1

1 Introduction The challenge of achieving both innovation1 and operational effectiveness2 within the same organisation is something that has puzzled both researchers and practitioners for decades. This challenge is particularly evident in research and development (R&D) departments where the research part of R&D often engages more extensively in exploratory activities3 in terms of generating new knowledge and ideas, and the development part of R&D engages more often in exploitatory activities4

Garcia et al., 2003

, i.e. realising existing ideas using available knowledge. This is not to say that exploration is equal to research or that exploitation is equal to development as some scholars suggest ( ), but rather that both the long-term and short-term perspectives of innovation are evident within the same organisation. The field of organisational ambidexterity5

O'Reilly & Tushman, 2013

emerged as a response to the challenges faced by organisations to manage both innovation and operational effectiveness ( ). This particular research field addresses the dual achievement of exploration and exploitation.

1.1 Challenges in the field of organisational ambidexterity

One approach to ambidexterity is to try to achieve both exploration and exploitation simultaneously within a single organisational unit without separating the two activities in time or space (Birkinshaw & Gibson, 2004). These organisations encounter specific challenges related to both exploration and exploitation, as well as challenges related to combining the two. Two specific challenges related to the field of organisational ambidexterity are addressed in this thesis. First, one shortcoming in the existing literature is that, although studies on the individual and team level exist, the majority of research focuses on the organisational level as 1 The outcome or process of realizing a creative idea, resulting in, for example, new products, services or work methods (Amabile, 1998) 2 The capability to satisfy today’s customers’ demands in terms of function, price, time, quantity, and place (Boer & Gertsen, 2003). 3 Activities with the aim of gathering knowledge, generating new knowledge, or coming up with new ideas, for example, search, experimentation, information gathering, and creative problem solving. 4 Activities with the aim of realising existing ideas using existing knowledge. 5 I define ambidexterity as the ability to achieve both exploratory output and exploitatory output.

2

the unit of analysis (Birkinshaw & Gupta, 2013). This makes it hard to generate actionable knowledge to help employees who focus mainly on either explorative activities or exploitative activities to become ambidextrous. Perhaps this is also one of the reasons why there is a lack of advice for how to accomplish ambidexterity at the operational level (Boer & Gertsen, 2003).

The second challenge is related to strategies that many companies have used in order to improve their operational effectiveness. Partially due to the success of Japanese firms, many organisations turned to process-management approaches6 Ohno, 1982, such as Lean ( ; Womack & Jones, 2003) and Total Quality Management (Hackman & Wageman, 1995), as a way to achieve operational effectiveness and high levels of flow efficiency. These ways of working have given promising results on the exploitation side of ambidexterity, but research has shown that it is less clear what effects that such ways of working will have on creativity and long-term innovation in an organisation (Benner & Tushman, 2002; Benner & Tushman, 2003). The work methods that are related to many process-management approaches give ample support for how to adhere to and improve on existing processes, but they give less support in how to fully renew processes or renew the organisation in times of turbulence in the environment (Benner & Tushman, 2002; Sitkin et al., 1994).

1.2 Aim of thesis and research setting The challenges described above identify the aim of this thesis, which is to explore the challenges related to innovation that employees encounter when working in an ambidextrous organisation where employees are expected to achieve both exploration and exploitation as an inherent part of their work. More specifically, this thesis seeks to understand how employees experience their abilities to engage in innovation while under the influence of exploitation-related incentives, efficiency targets, and follow-up measures that can be found in many process-management approaches.

The main case company in this thesis is the Swedish-based company Scania, a developer and manufacturer of heavy trucks, buses, and engines. The research setting was that of an industrial PhD project, meaning that the PhD student was employed within the company that

6 Management philosophies with the aim of making organizational processes more efficient by mapping, adhering to, and improving the processes (Benner & Tushman, 2003).

3

serves as the main case (Kihlander et al., 2011). The industrial PhD setting has enabled in-depth studies of everyday operations of the organisation, as well as access to the operational level that is usually difficult to achieve in research on organisations. The closeness to the organisation and the involvement of a number of senior managers in the PhD project has enabled continuous validation of the findings. The PhD student has, during the whole project, been affiliated with the unit of Integrated Product Development at KTH Royal Institute of Technology. This combination of closeness to both practice and academia has served to ensure both the academic and practical relevance of the thesis. In addition to Scania, five other organisations with R&D departments in Sweden have permitted the use of surveys and interviews for additional data collection.

1.3 Theoretical positioning This thesis positions itself within the field of organisational ambidexterity. Organisational ambidexterity is a complex area that spans a number of research fields such as innovation, organisational learning, and management (Raisch & Birkinshaw, 2008). For that reason, this thesis also needs to span a large number of theoretical areas. It builds primarily on theory from creativity, operations management, and project management, and its findings contribute to subsets of these research fields (Figure 1). The research scope in the thesis is limited to the micro-level, that is, the individual level of ambidexterity, and thereby addresses a perspective that has gained relatively little attention compared to the organisational level (Birkinshaw & Gupta, 2013).

4

Figure 1: Areas of relevance and contribution to this thesis with a main theme of micro-level contextual ambidexterity (model adopted from Blessing & Chakrabarti, 2009, p. 65). Areas of relevance are represented by the grey areas with arrows going into the

black circle. Each area of relevance is in turn made up of the areas written in the smaller circles that overlap the grey areas. Areas to which this thesis contributes are

marked by dashed arrows pointing at them.

1.4 Structure of the thesis This thesis builds on six appended papers, each addressing different perspectives of combining exploration and exploitation in R&D departments that have adopted process-management approaches. The findings in each paper have been aggregated and further analysed in this summary in which a synthesis is presented alongside a description of the research’s contribution to theory and practice. The following sections are included in the summary: a theoretical exposition including research questions, a description of the research methods and the motivation for methodological choices, a summary of the appended papers, a summary and analysis of the main findings, and lastly, a discussion of the implications for theory, practice, and future research.

5

2 Achieving organisational ambidexterity

This chapter presents the theory in the field and problematises the tensions between exploration and exploitation that is at the heart of challenges in ambidexterity research. It starts with a brief account of the origin of ambidexterity research and then goes into more depth concerning exploration and exploitation and their relationships to innovation and process management. The main part of the chapter describes the combination of exploration and exploitation and different ways that organisations can achieve ambidexterity. The final part of this chapter focuses on contextual ambidexterity and the implications it has at the micro-level. The theory chapter ends by presenting the two research questions that this thesis seeks to answer.

2.1 Dual perspectives in R&D Innovation speed became a focus question of R&D management in the late 1980s and early 1990s (Kessler & Chakrabarti, 1996; Rothwell, 1994). The ability of Japanese firms, in particular Toyota, to achieve operational effectiveness motivated many scholars and organisations to try to uncover the secrets hidden in their ways of working. Many publications written in the 1990s and 2000s dealt with what was dubbed The Toyota Way (Liker, 2004), Toyota Production System (TPS) (Ohno, 1982), or simply Lean production (Krafcik, 1988). A few years later, the process-management approach that had proven so successful in production was followed by related literature that focused on the implementation of Lean in R&D and services (e.g. Liker & Morgan, 2006). Increased competition called for an all-around operational effectiveness where all of the operations of a company were aimed at delivering products to their current customers in a way that they found satisfactory in terms of price, time, and function (Boer & Gertsen, 2003). Thus, many R&D departments came to adopt process-management approaches in order to improve their processes so that they could get new products to the market faster and at a lower price than their competitors could.

6

2.1.1 From operational effectiveness to innovativeness During the 1980s, product-development companies were also faced with the challenge of combining operational effectiveness with high innovativeness (Wheelwright, 1992). In addition, customers became increasingly knowledgeable and demanding, while at the same time more and more advanced technology became readily available that could be integrated into many products (Clark & Fujimoto, 1991). Being able to offer innovative products became a key component of a successful organisation (Boer & Gertsen, 2003; Bolwijn & Kumpe, 1990)

Since then, the R&D management model has shifted over to a focus on timely completion of product development and predictability in projects as well as on loosely tied networks for co-innovation (Nobelius, 2002; Nobelius, 2004). Nevertheless, a shift in focus of organisational development in R&D does not mean that past foci will go out of fashion. R&D organisations of today must still be successful in the types of activities that characterised earlier generations of R&D management (Nobelius, 2002). Thus, the new competitive landscape calls for organisations to achieve both exploration, such as exploring new ideas and opportunities, and exploitation in the sense of turning ideas into products and services in an efficient and effective way. Organisations that achieve these dual perspectives are what we call ambidextrous organisations.

2.2 The ambidextrous organisation As a logical consequence of the focus on operational effectiveness in the field of R&D management, researchers started to study the consequences that a dual focus on effectiveness and efficiency could have on aspects of product development such as experimentation, creativity, and learning. This research stream generally came to be characterised by the term ambidexterity, meaning the ability to manage two different things, and in this case this means achieving both exploration and exploitation (O'Reilly & Tushman, 2004). Although ambidexterity had been mentioned in earlier publications (e.g. Duncan, 1976), two seminal pieces are often credited as the starting point of ambidexterity as a research stream. First, March (1991) presented the challenges of combining the two perspectives of exploration and exploitation and the role those challenges play in organisational learning. Exploration was presented as a collective term for things like risk taking, innovation, flexibility, and discovery, and exploitation included things

7

like refinement, efficiency, implementation, and execution. Secondly, Tushman and O’Reilly (1996) argued that the organizational design in which the combination of exploration and exploitation could be most successfully managed was the ambidextrous organisation. In Tushman and O’Reilly’s research, the ambidextrous organisation separates its exploratory and exploitatory units while keeping a tight link between the two on the senior management level.

The definitions of exploration, exploitation, and ambidexterity differ depending on the author, but most often ambidexterity is defined as the successful combination of exploration and exploitation. The type of organisation that Tushman and O’Reilly held forward as the ambidextrous organisation was later challenged as being merely one out of several possible ways to achieve both exploration and exploitation.

2.2.1 Key concepts of ambidexterity The two concepts of exploration and exploitation are central to the field of organisational ambidexterity. The vast majority of recent publications in top journals in the field use the words “exploration” and “exploitation” to define ambidexterity (Birkinshaw & Gupta, 2013; Simsek, 2009). However, the field lacks consensus on how the studied concepts should be operationalised, specifically how exploration and exploitation should be measured (Birkinshaw & Gupta, 2013). Many publications also lack definitions of exploration and exploitation, or they refer to the definitions given by March (1991), which in turn suffer from being rather ambiguous and broad. For example, can innovation, which March accredits to exploration, be accomplished without any level of implementation, which he accredits to exploitation? The next part of the thesis presents the two concepts, problematises their conceptualisation, and presents the definitions used in this thesis.

2.3 Exploration activities – A key ingredient of R&D work

The definition of exploration that is commonly referred to among ambidexterity scholars states that “Exploration includes things captured by terms such as search, variation, risk taking, experimentation, play, flexibility, discovery, innovation.” (March, 1991, p.71). The definition is broad, especially given the addition of “things captured by terms such as”. Can, for example, concepts like creativity and ideation be considered to be captured by the term innovation? Does it also include, as

8

mentioned above, implementation of creative ideas, which is commonly held forward as the very definition of innovation (Amabile et al., 1996)? How, in that case, does implementation fit within the description of exploitation? Despite these questions, March’s definition of exploration does give us a picture of what types of activities constitute exploration. The definition becomes even clearer when contrasted to the things he puts forward as being examples of exploitation, and perhaps that is why his definition has become so widely used.

Some scholars have proposed other definitions for the exploration side of ambidexterity. Adler (1999), for example, discusses flexibility and contrasts it with efficiency, whereas Gibson and Birkinshaw (2004) address adaptability and use the term “alignment” instead of exploitation. Yet others put forward innovation (or certain types of innovations) as the exploration side of ambidexterity, e.g. disruptive innovation as opposed to sustaining innovation (Danneels, 2006), discontinuous innovation as opposed to incremental innovation (Tushman & O'Reilly, 1996), and innovation as opposed to efficiency (Sarkees & Hulland, 2009). Innovation is a key concept of this thesis alongside exploration, exploitation, and ambidexterity. It is, therefore, relevant to analyse the implications of equating innovation with exploration.

2.3.1 Innovation and exploration The basic definition of innovation (which will be used in this thesis) states that innovation is the successful implementation of a creative idea (Amabile et al., 1996). A creative idea, in turn, is defined as an idea that is novel and useful (Amabile et al., 1996). While those definitions are suitable to ensure a basic common ground, they are far from satisfactory if one wishes to fully understand creativity and innovation in an organisational context. To begin with, the meanings of “novel” and “useful” are not universal. Some argue that a creative idea should be novel and useful for the context in which the idea was generated, whereas some argue that an idea is creative only if it is new to the world (see Garcia & Calantone, 2002 for a review). Likewise, the meaning of the word “successful” is not universal and can, for example, include or exclude market success and company profit. In addition, the realisation process often contains a multitude of activities such as further development of the idea, testing, marketing initiatives, and ensuring high

9

quality in the production process, and these are in turn governed by factors like management and organisational culture.

Innovation can thus be considered a multifaceted construct even though it has often been measured as a single construct. More and more researchers are, however, arguing that it must be seen as a compound construct where its many facets are taken into consideration (Shalley et al., 2004; Woodman et al., 1993).

Phases and activities of innovation In a comprehensive review of the literature about innovation, Crossan and Apaydin (2010) summarised research on the innovation process into a condensed model. In that model, the innovation process contains phases of initiation and decision-making, development, implementation, and commercialisation (Figure 2). The initiation phase includes aspects like attitudes towards new ideas, the generation or adoption of new ideas, and concept development. The development phase includes problem solving, design, and detailed development of concepts. The implementation phase includes both validation of concepts and a handover to production as well as the actual production of the product. Finally, the commercialisation phase includes marketing activities such as market research, market tests, and post-launch reviews (Crossan & Apaydin, 2010) (Figure 2). In addition to the phases of innovation, taking a process perspective of innovation means that project management and portfolio management need to be taken into consideration (Crossan & Apaydin, 2010).

Figure 2: A model of the innovation process, inspired by Crossan and Apaydin (2010).

The red in the model marks the phases where the concentration of exploration activities is the highest. The activities outlined in the different phases of innovation

will likely overlap each other, but for the sake of visualisation they have been depicted as a linear and sequential process.

2.3.2 Exploration activities Figure 2 gives a view of what an innovation process can look like. This innovation process arguably contains elements of exploitation as defined by March’s proponents, whereas it is the very definition of exploration for those who equate exploration with innovation. In order to avoid many of the pitfalls of the innovation construct, while at the same time narrowing the definitions down to something simple enough to describe,

10

this thesis will focus on the concept of “exploration activities”. Exploration activities are defined as “activities with the aim of gathering knowledge, generating new knowledge or coming up with new ideas.” Examples of exploration activities are, for example, identifying opportunities of innovation, ideation, and the development of concepts (Bertels et al., 2008; Holmen et al., 2007), but also experimentation, search, and discovery in a more general sense (March, 1991). These activities are exploratory regardless of which product development phase they take place in, although the highest concentration of exploration activities can be found in the first phases of the innovation process (Figure 2).

This definition entails that employees, regardless of which phase of the innovation process they are working with, are likely engaged in both exploration activities and exploitation activities. This view of exploration stands in contrast to other commonly used definitions such as those relating to innovation that are mentioned above. The front end of innovation undoubtedly includes aspects of exploration such as creativity and experimentation (Florén & Frishammar, 2012), but if one would equate the front end with exploration one would neglect the difference in nature of the activities that are performed during that phase. In an R&D department, it is probable that the same employee who runs an experiment and thereby gains new knowledge (exploratory activity) will also write the report documenting the knowledge that was gained (exploitatory activity).

The definition of exploratory activities in this thesis is thus grounded in the idea that the early phases of innovation often include exploitative activities just as the later phases often include exploratory activities (Rosing et al., 2011). Studying exploration activities as defined in this thesis enables the inclusion of phases that share the same need for exploration but that are not part of a formal product development process, such as skunk work or early advanced engineering. It also enables the inclusion of activities late in the innovation process that require exploration, for example, the redesign of a product as a response to deviations discovered when the first parts are produced. Similarly, it excludes activities in the early phases of product development that are not explorative in nature such as many administrative tasks.

11

Multi-project environments To complicate things further, employees in an R&D environment are often involved in several projects simultaneously. These conditions often come with challenges in allocating enough resources to each project and require reactive behaviour in order to solve problems that occur as a result of resource shortages (Engwall & Jerbrant, 2003). Employees might work with the implementation phase of one on-going project at the same time as the initiation of the next product-development project (Figure 3). This means that employees must switch between projects with a high concentration of exploration activities and projects with a focus on execution with the aim of launching a new product in a timely manner, i.e. a high concentration of exploitative activities. Employees as well as project managers must be able to shift between explorative and exploitative activities and the mindsets related to each activity (Bledow et al., 2009). Furthermore, those with control over the project portfolio and resource assignment need to be able to prioritise between projects of different maturity without being overly biased towards exploitative projects (Levinthal & March, 1993; March, 1991).

Figure 3: A multi-project view of R&D activities. At any g iven point in time, an

employee might be involved in both early-phase projects with a high concentration of exploration activities and late-phase projects with a high concentration of exploitative

activities.

2.4 Achieving efficient exploitation The other side of ambidexterity addresses exploitation. As in the case of exploration, many scholars rely on March’s description when defining exploitation. March states that “Exploitation includes such things as refinement, choice, production, efficiency, selection, implementation, execution.” (March, 1991, p.71). Although widely used, this definition also comes with some issues. Most words in the definition describe something you do, such as “production”. March, however, includes the word “efficiency”, which leads to the question of whether all production can be considered exploitation or if only efficient production processes

12

qualify. Other scholars have, for example, chosen to concentrate particularly on the efficiency aspect of exploitation and contrasted that with flexibility (Adler et al., 1999). Another issue with March’s definition is that refinement, understood as changing one way of working for one that is better, often requires searching for new ways of doing things as well as variation from the original way of working. Both search and variation are part of March’s (1991) definition of exploration.

Exploitation can alternatively be defined in terms of alignment contrasted with adaptability (Gibson & Birkinshaw, 2004). Alignment means that different activities in a business are organised in order to work towards a shared goal. Ambidextrous organisations would be required not only to align, but also to adapt, i.e., to reconfigure their activities to respond to changes in the environment (Gibson & Birkinshaw, 2004). This thesis uses a conceptualization of exploitation that is more aligned with the notion of knowledge and competencies. Janssen and colleagues (2005), for example, define exploitative innovation as innovation that builds on existing knowledge and skills or that are refinements of existing products. This definition, however, becomes problematic in the light of the definition of innovation that is used in this thesis, namely that innovation is the realisation of an idea that is novel and useful (Amabile et al., 1996). The novelty dimension implies that innovation cannot fully be based on existing ideas, but that new ideas need to be generated. A similar definition, one that does not suffer from the same issue, describes exploitation as exploiting existing competencies (Lubatkin et al., 2006). This way of defining exploitation does not carry the issue of building on a compound concept, such as innovation, that in turn arguably depends on both exploration and exploitation. This thesis will use a definition of exploitation that is similar to that of Lubatkin et al. (2006), namely that exploration activities are activities with the aim of realising existing ideas using existing knowledge.

2.4.1 Process management as a means to achieve efficiency Another aspect of exploitation that is of relevance to this thesis is the role that efficiency plays. For a company seeking profit, it is reasonable to want to strive for efficiency in instances where both the idea and the knowledge required to realise the idea are available, such as during exploitation activities. For that reason, efficiency is implicitly related to exploitation. Furthermore, the aim of this thesis is to understand the

13

employee situation in ambidextrous contexts, specifically in organisations where there is a pressure to make processes more efficient by means of process-management approaches.

Common traits of process management Process management can be interpreted as an umbrella under which we collect management philosophies aiming for increased efficiency and effectiveness by means of mapping, improving, and adhering to processes (Benner & Tushman, 2003). The promise of process management paradigms is in most cases improved efficiency, quality, and total cost, and process management is the means by which many organisations try to achieve operational effectiveness (Benner & Tushman, 2003). We often find these approaches in manufacturing or other highly repetitive processes. Over the years, however, process-management methods have, to an increasing extent, been introduced in other types of operations, such as R&D. There are, however, many instances where this transition has been made without fully acknowledging differences in uncertainty and repetitiveness and where such initiatives have consequently failed (Sitkin et al., 1994). Some of the most well-known process-management approaches of today are Total Quality Management (TQM) (Hackman & Wageman, 1995), Six sigma (Harry & Schroeder, 2006), Lean (Womack & Jones, 2003), Business Process Reengineering (BPR) (Davenport & Short, 1990; Hammer & Champy, 1993), and Agile (Highsmith & Cockburn, 2001). In addition to the process-management paradigms described by their advocates, countless variants exist that are the result of adopting and adapting process-management approaches to specific contexts. For example, Lean supply chain (Lamming, 1996) can be mentioned, as well as the trend of applying Lean thinking in healthcare resulting in a paradigm of Lean healthcare (Waring & Bishop, 2010).

Mapping, adhering to, and improving processes Roughly speaking, process management approaches have in common that they seek to reach efficiency by mapping, adhering to, and improving processes (Benner & Tushman, 2003). The approaches all seek efficiency and effectiveness, but the emphasis in each process-management approach differs. Lean, for example, stresses the maximising of customer value through flow efficiency (Modig & Åhlström, 2012; Womack & Jones, 2003), whereas Six sigma puts more emphasis on reducing variability in processes through an extensive use of

14

metrics (Schroeder et al., 2008). While Six sigma forms separate improvement projects with trained leaders, TQM and Lean seek to make process improvement every employee’s job.

What the process-management approaches share is that they generally strive to reduce variation and to increase efficiency (Benner & Tushman, 2003). Efficiency, in the field of process management, often refers to flow efficiency rather than resource efficiency, although resource efficiency is also an important aspect. Lean, for example, emphasises waste reduction as a key component. Waste in R&D processes typically includes unnecessarily waiting for decisions or information, corrections of mistakes that could have been foreseen, and bottlenecks caused by the overutilization of certain resources (Morgan & Liker, 2006). Elimination of waste and optimising flow are closely linked within Lean. While many would intuitively point out idle employees as a type of waste, a certain overcapacity of human resources is actually necessary in a process optimised on flow and when aiming to keep queues to a minimum (Erlang, 1909; Little, 1961; Reinertsen, 2009).

Assumptions in process management Efficiency implies that what is to be produced is known, and increasing flow or resource efficiency is difficult if we do not know what we want to gain more of. Lean even stipulates defining what constitutes customer value as its first step (Womack & Jones, 2003) and Six sigma uses metrics to confirm that the process improvements did indeed improve performance (Schroeder et al., 2008). Because the mission of R&D is oftentimes to define what constitutes value for future customers, process management is not as easily applicable as it is in production. Two other important assumptions are also needed in order to ensure the value of mapping processes, namely that the process can be captured in words, that is, externalised (Nonaka, 1994), and that there is value in finding a best practice, i.e. the situation in which the process is used is recurring. These two assumptions are not always fulfilled in activities carried out in R&D departments. The creative process is difficult to externalise, and although many steps in the innovation process are recurring, they often differ slightly between repetitions. Some scholars even argue that highly customized service processes should not be standardised (Hall & Johnson, 2009). The next section further analyses these and other challenges of applying process management as a means to improve exploitation in R&D.

15

2.4.2 Challenges of applying process management in R&D Applying process management in R&D has spurred a debate as to whether it can be harmful for exploratory activities (Benner & Tushman, 2003). Because R&D work typically includes the development of innovations, it must rely on some aspect of exploratory activities. Some scholars argue that process management will harm innovation, in particular architectural innovation and radical innovation (Benner & Tushman, 2003), and that it will lead to a shortage of money for product development (Cusumano, 1994). In contrast to such claims, others (Adler et al., 2009) (Adler et al., 2009) argue that one of the secrets behind Toyota’s success is the fact that they use the same type of problem-solving methods for continuous improvement as they do for developing radical innovations. Furthermore, survey-based research on the different process-management paradigms has shown that the majority of R&D engineers experience positive effects or no effects on their creative behaviour from adopting process management-related approaches such as TQM, Kaizen, Just-in-time, and Lean production (Ekvall, 2000). While the harmful effects that process management might have on R&D work can be debated, there are two aspects, described below, that are particularly relevant in this thesis.

Combining variability and risk taking with predictability In production, the outcome of a process should be the same in each cycle and the process should undergo slow evolutionary improvement. In R&D, however, the outcome must change from project to project because no value is created if the next-generation product specification is identical to the previous one (Reinertsen & Shaeffer, 2005). Variability in production is considered waste because product quality is defined according to the product specification. Any deviation from that specification is a quality flaw. In R&D, however, variability is essential to creating value, and instead we should differentiate between good and bad variability. Good variability is connected to the uncertainty that comes with trying something new, which in turn is one key to producing innovative output. In contrast, bad variability includes aspects like not adhering to processes and thereby producing rework that does not come with any learning opportunities (Reinertsen & Shaeffer, 2005). Because cycle times in R&D are longer, it is possible that advances in technology and process development have been made since the previous cycle, and these motivate larger changes in process.

16

Furthermore, some principles of process management, such as the pull principle in Lean (Womack & Jones, 2003), build on the assumption that it is possible to predict customer demand with the same or even longer foresight than the lead times. This is oftentimes not the case in innovation processes. The cost of keeping stock is another aspect that differs between contexts. In markets where demand is uncertain, margins are high, and costs of over-production are relatively low, a certain level of over-production can be economically valid (Fisher et al., 1994).

Bias towards short-term gains The second aspect often considered problematic when introducing Lean in R&D is that of short-term thinking. Aiming for efficiency and predictability might in some instances undermine the ability to adapt to changes in the environment (Levinthal & March, 1993). Furthermore, the tendency towards predictability in processes creates less flexibility in the projects, which might in turn threaten innovation (Richtnér & Åhlström, 2010). The aspect of bias towards short-term gains highlights one of the main challenges when both exploration and exploitation are combined within an organization. Balancing short-term gains with long-term opportunities is at the very heart of ambidexterity.

2.5 Combining exploration and exploitation While achieving exploration and exploitation separately does come with its own challenges, it is the combination of the two that stands at the centre of the ambidexterity literature. Different organisations use different strategies in order to achieve this, and these strategies have implications for, among other things, organisational design, coordination strategies, and management practices.

2.5.1 Four strategies for achieving ambidexterity The following four different types of ambidextrous organisations can be discerned in the ambidexterity literature: structural ambidexterity, that is, organisational separation; ambidexterity by temporal separation; ambidexterity by domain separation; and contextual ambidexterity where both exploration and exploitation are achieved within one organisation at a given point in time (Lavie et al., 2010).

Structural ambidexterity One of the most common views of ambidexterity is that of structural ambidexterity presented by Tushman and O’Reilly (1996). In the

17

structural ambidexterity approach, organisational ambidexterity is realised by creating two separate units, one for exploration and the other for exploitation. These two units should be separated geographically as well as culturally but should be tightly connected on the senior management level (O'Reilly & Tushman, 2004; Tushman & O'Reilly, 1996). A structurally ambidextrous organisation will, however, likely encounter challenges with integration issues that arise when two organisations with little cultural and methodological overlap need to collaborate as a project moves from the exploration phase to the exploitation phase (Lawrence & Lorsch, 1967).

Temporal separation In the view of ambidexterity that suggests temporal separation, an organisation cyclically shifts between periods of exploration and periods of exploitation in order to become an ambidextrous organisation. Temporal separation as a means to achieve ambidexterity springs from the idea of punctuated equilibrium, that is, long periods of stable conditions that are disrupted by shorter periods of turbulence in which the organisation adapts to the new conditions (Gersick, 1991; Romanelli & Tushman, 1994; Tushman & Anderson, 1986). Many of the challenges of temporal separation are related to managing the transition periods and identifying when there is a need for a transition. During those periods, the organisation will need to overcome the inertia embedded in having done the same thing for a long period of time (Leonard-Barton, 1992; Levinthal & March, 1993). Thus, while temporal separation might enable an organisation to pursue both exploration and exploitation, albeit not simultaneously, it does not mean that the organisation does not have to handle the trade-offs between exploration and exploitation during the transition periods (Lavie et al., 2010). At the point where an organisation faces a transition period (a punctuation), the employees will have worked under stable conditions for a long period of time, which will lead to decreased levels of preparedness for change (Levinthal & March, 1993). A possible remedy for this problem is to keep the organisation on its toes by means of “deliberate perturbations”. Deliberate perturbations means that the organisation deliberately induces stimuli that force the organisation to discover new routines (Brunner et al., 2010). That way, organisations can keep their exploitation at high levels and still maintain their preparedness for change at high levels so that it can act quickly when externally induced punctuations occur (Figure 4).

18

Figure 4: The difference in organisation’s preparedness for change between the model of punctuated equilibrium (Tushman & Anderson, 1986) and the model of deliberate

perturbations (Brunner et al., 2010).

Contextual ambidexterity Contextual ambidexterity differs from the other three types of ambidexterity in that it does not build on separation or buffering between exploration and exploitation (Birkinshaw & Gibson, 2004). Instead, a dual engagement in exploration and exploitation is achieved on the individual or group level and thereby makes the organisation ambidextrous (Lavie et al., 2010). Proponents of the other models often hold forward the inferiority of the contextual model because a contextually ambidextrous organisation will have to deal with trade-offs of simultaneously engaging in exploratory and exploitatory activities. Such trade-offs include limits in resource allocation and the focus on, and communication of, different types of organisational outcomes (March, 1991). Scholars focusing on contextual ambidexterity acknowledge that it is difficult, if not impossible, to fully bypass the trade-offs inherent in exploring and exploiting within the same organisation (Gibson & Birkinshaw, 2004). There are, however, some conditions under which the contextual ambidexterity model has advantages that the other models lack. For example, the cost of integrating something that has once been separated, as in the structural ambidexterity approach, can be particularly high for certain organisations (Bledow et al., 2009). In addition, learning opportunities between people working with exploratory and exploitatory activities can be lost if they are separated (Bledow et al., 2009).

Domain separation The third and most recently presented way of achieving ambidexterity is by domain separation. This type of ambidexterity focuses mainly on ambidexterity through alliances with network partners. The idea is that

19

an organisation chooses whether they are to be explorative or exploitative in a number of domains. These domains consist of the function domain (generating or exploiting knowledge within the alliance), the structure domain (building alliances with new or existing partners), and the attribute domain (building alliances with partners whose attributes are similar to or different from those of previous partners) (Lavie & Rosenkopf, 2006). The function domain is perhaps the domain in which this way of viewing ambidexterity differs the most compared to contextual ambidexterity. Proponents of the contextual ambidexterity approach argue that knowledge can be created and exploited within a certain function domain (Birkinshaw & Gibson, 2004), whereas proponents of the domain separation approach argue that knowledge creation and exploitation are two extremes on a continuum and that they cannot be combined within the same function domain (Lavie & Rosenkopf, 2006). The domain separation approach also encounter challenges similar to the structural ambidexterity model when it comes to integrating units that have once been differentiated in order to enhance either exploration or exploitation (Lawrence & Lorsch, 1967).

Summary of ambidexterity perspectives This thesis focuses mainly on contextual ambidexterity. It should, however, be pointed out that although most research focuses on one of the ambidexterity strategies, many organisations will, in practice, adopt a mix of the different approaches (Kauppila, 2010; O'Reilly & Tushman, 2013). Table 1 presents a summary of the different strategies.

20

Table 1. A summary of the different perspectives of ambidexterity (inspired by Lavie et al., 2010).

Structural separation Temporal separation Contextual ambidexterity Domain separation

Des

crip

tion

Enhancement of exploration and exploitation is enabled through separation of exploration into a separate unit with separate rules and a separate culture specifically adopted to support exploration. Strong linkages on top management level.

The engagement in either exploration or exploitation follows the conditions in the corporate environment. During turbulent times, focus lies on exploration and during stable times focus is on exploitation.

The dual achievement of both exploration and exploitation within an organisation.

Focus on exploration in some alliances, such as engaging with new suppliers, and focus on exploitation in other alliances, such as working with partners that are similar to the organisation.

Figu

re

Adv

anta

ges

Employees can focus on either exploration or exploration. Mechanisms can be aligned to support one single focus.

Employees can focus on either exploration or exploitation in any given point in time.

All business units are kept alert and trained in both exploration and exploitation. Deep expert knowledge can be built. The history behind decisions in the early phases is known in the later phases.

Every alliance can focus on either exploration or exploitation whereas both can be achieved on a company level.

Dis

adva

ntag

es

Difficulties in handover between the two organisations. Double sets of expert knowledge needed.

Organisation is not prepared for change when changes in the environment occur.

Requires balancing between the two perspectives on a group or individual level. A risk of applying exploitation-enhancing thinking on exploration and vice versa.

Pockets of the organisation will have little experience from exploration. Handover from alliance to organisation in the industrialisation process can be troublesome.

2.5.2 Other ways of differentiating ambidexterity research In addition to the four ways of achieving ambidexterity described above, there are several other ways in which one can study the phenomenon. First of all, we can choose to either take on a static approach or a dynamic approach. It has been argued that to evolve the research field inquiry must move from studying whether organisations have achieved ambidexterity or not to studying what the process of achieving and

21

maintaining ambidexterity looks like (Gupta et al., 2006; Simsek, 2009). Secondly, we can study the phenomenon at different levels of analysis. The organisational level is perhaps the most common one, but research on that level struggles to produce actionable knowledge that can help organisations work their way toward ambidexterity. Several scholars argue that there is a need for more research in the field that focuses on gaining actionable knowledge of how to achieve ambidexterity7

Boer & Gertsen, 2003 on the

operational level ( ; Martini et al., 2013; O'Reilly & Tushman, 2011) as well as research that addresses multiple levels of analysis in the same study.

Irrespective of the way we study ambidexterity, and what approach an individual or firm uses to achieve it, the challenge remains highly relevant and still requires more research (Birkinshaw & Gupta, 2013; O'Reilly & Tushman, 2013). There is a need to look beyond exploration and exploitation as antithetical, or as dichotomies where the two are mutually exclusive, and instead study it as a duality of two things that can co-exist at the same time (Andriopoulos & Lewis, 2010; Magnusson & Martini, 2008). The ways in which an organisation can engage in exploration and exploitation has been summarized by Boer & Gertsen (2003), who argue that the only organisational form that truly manages to combine both exploration and exploitation is the dual organisation (Figure 5). The ambidexterity approach described above that builds on temporal separation would fit better in the classification of the binary organisations that have the capability to shift between modes of exploration and exploitation, which are referred to in this model as mechanistic and organic forms of organising (Boer & Gertsen, 2003; Burns & Stalker, 1961).

7 Boer and Gertsen (2003) use the term Continuous innovation, rather than ambidexterity, as the ability to combine operational effectiveness (exploitation) with strategic flexibility (exploration).

22

Figure 5: A mapping of the different forms of organisations based on their ability to work with exploration and exploitation. (Figure adapted from Boer & Gertsen, 2003,

p.810).

2.6 The paradoxical relationship between exploration and exploitation

The ability to manage both exploration and exploitation within the same company is challenging given the difference in nature and aim of the two ways of working. Many scholars would even argue that exploration and exploitation are paradoxical in nature, contradictory yet mutually interdependent. An illustration of this paradoxical nature can be found in a review of the ambidexterity literature in which the authors claim that exploration and exploitation are inherently antithetical: when exploration increases, exploitation inevitably decreases (Lavie et al., 2010). Figure 6 is an adaptation of their illustration of the relationship between exploration and exploitation. The figure shows that exploitation is necessary to generate income for the company, income that is in turn necessary in order to cover the expenses of exploration. Exploration on the other hand will lead to new opportunities, opportunities that are in turn necessary in order to have anything to exploit. Because exploration and exploitation are antithetical (according to Lavie et al., 2010), the paradox is a fact. While this is a pedagogical illustration of the challenges of combining exploration and exploitation, one can criticise the assumption of antithetical relationship between the two constructs. The paradoxical nature of ambidexterity is dependent on the choice of definition. If, as is the case in the work of Lavie et al. (2010), one defines exploration and exploitation as the two extremes of a single continuum, the paradox is inherent in the definitions. If one instead takes a view of exploration and

23

exploitation as interdependent measures, the paradox is no longer inherent in the conceptualisation, but might still be present in practice. Given that the challenge of ambidexterity is relevant only in its practical embodiment (theories do not need to make money), the definition of the two on a single continuum seems overly strict and theoretically driven.

Figure 6: An illustration of the paradoxical relationship between exploration and

exploitation (adapted from figure 2 in Lavie et al., 2010, p.117).

2.6.1 Disentangling the ambidexterity paradox The four strategies for achieving ambidexterity that are described in the previous sections are all examples of attempts to bypass, and sometimes resolve, the ambidexterity paradox. By using principles of handling paradoxes, such as separating in time or organisational structure (Poole & Van de Ven, 1989), exploration and exploitation can co-exist without outcompeting each other. The contextually ambidextrous approach, however, does not build on separation, but rather builds on creating a culture and management approach that allow and encourage employees to combine exploration and exploitation.

Another approach to disentangling the ambidexterity paradox is to consider whether exploration and exploitation really should be considered as two separate constructs, albeit difficult to manage at the same time, or as two ends of a single continuum. The latter definition manifests the paradoxical relationship in the definition itself. Or, to use the words of Birkinshaw and Gupta, it “ends up defining away the interesting parts of the study, that is, the ability of firms to deliver on both dimensions at the same time” (Birkinshaw & Gupta, 2013, p. 296). The way the concept of ambidexterity is defined differs among studies and researchers. On this note, it is worth pointing out that Tushman, who is well established in the field, changed position from defining exploration as “opposing activities along a continuum” in 2010 (Lavie et al., 2010) to defining it three years later as two separate constructs (O'Reilly & Tushman, 2013).

24

Part of the explanation behind this split in definitions can probably be found in the operationalisation and measure of exploration and exploitation in research studies. In questionnaires addressing balance, for example, the logical conceptualisation is that of a single scale, and in studies addressing the two constructs separately, the operationalisation will also have to be as separate constructs. The simultaneous achievement of exploration and exploitation is considered a desirable goal whether or not researchers believe it is plausible. From that logic, it is reasonable to conclude that the definitions of exploration and exploitation should avoid contributing to manifesting the paradoxical relationship (Magnusson & Martini, 2008). Whether or not these two constructs are impossible to achieve simultaneously can then be answered by empirical data.

Activity or output The second thing to consider is how exploration and exploitation should be combined. Many of the definitions of ambidexterity include words like “pursuing exploration and exploitation simultaneously” (Jansen et al., 2008), “conduct two paradoxical things at the same time” (Im & Rai, 2008), and “the inherent trade-offs between exploration and exploitation reinforce their operationalisation as opposing activities along a continuum” (Lavie et al., 2010). These types of definitions of ambidexterity imply that it is the simultaneous conducting of activities of an exploratory and exploitatory nature that leads to a successful ambidextrous organisation, whereas what matters for a company’s long-term survival is its ability to produce outputs for both current and future customers. Failure to make a distinction between output and activities or processes is a common problem in innovation management literature because the word “innovation”, just like “exploration”, can describe both an output and an activity (Crossan & Apaydin, 2010). I suggest that this distinction between activities and output might hold one key to how the paradox of ambidexterity can be disentangled.

A new definition of ambidexterity Given the state of the literature in the field, this thesis will make use of a definition of ambidexterity that focuses on output rather than activities. While exploration activities and exploitation activities, as defined earlier, are important ingredients of an ambidextrous organisation, it is the organisation’s ability to produce exploratory and exploitatory output that ultimately determines its chances of success. In this thesis exploratory

25

output is defined as “new knowledge and new ideas”, whereas exploitatory output is defined as “realised artefacts or services that build on existing knowledge and ideas, as well as information manifested in the process towards realised artefacts and services, for example drawings and documentation”. Consequently, ambidexterity is defined as “the ability to achieve both exploratory and exploitatory output”.

2.7 The challenges and benefits of contextual ambidexterity

This study focuses specifically on contextual ambidexterity. Given its close links to contextual factors such as leader behaviour, support, trust, and ways of working (Gibson & Birkinshaw, 2004), the team and the individuals’ perceptions of their work environment are relevant units of analysis. In contrast, studies of the other types of ambidexterity often focus on the organisational level (Lavie et al., 2010). In the case of structural ambidexterity, the individual level would make little sense because individuals in such an organisation would work either in an exploratory or an exploitatory organisation.

A number of reasons support studying ambidexterity with a contextual approach. First, synergistic effects such as using ideas that emerge during exploitation as input to new exploration activities can be lost if exploration and exploitation are separated too rigidly (Bledow et al., 2009). In addition to a loss of synergy, models of ambidexterity that build on separation must take into account the cost of integration of their differentiated units (Lawrence & Lorsch, 1967). Third, it has been shown that business units and firms can manage both exploration and exploitation simultaneously (Gibson & Birkinshaw, 2004; He & Wong, 2004), and one can even argue that most individuals are required to combine some level of exploration and exploitation in their everyday work (Rosing et al., 2011). The requirement that individuals handle both exploration and exploitation is certainly true for R&D employees if one accepts the definitions of exploration and exploitation activities put forward in this thesis. Even if a person works with the early phases of concept development, the exploitatory aspects play a part in how existing knowledge is exploited in activities like report writing or test preparation. Whereas those activities might contain high levels of exploration as well, it is likely that there is an extensive source of pre-existing knowledge to exploit. Another example in R&D concerns the activities carried out in the later phases of development. In that case, the overall concept is often

26

decided and laid out but the details are often still not in place. Intense exploratory efforts are often needed to figure out, for example, what type of bracket will hold a part in place or what the optimal configuration of a wire harness will be.

2.7.1 Micro-level contextual ambidexterity Contextual ambidexterity is less researched compared to, for example, structural ambidexterity. Even less attention has been given to the individual level, that is, the work situation among the employees responsible for achieving ambidexterity (Turner et al., 2013). The literature that does exist suggests that ambidexterity at the individual level, termed here as “micro-level contextual ambidexterity”, depends on both individual characteristics and contextual factors (Raisch et al., 2009).

Characteristics of ambidextrous individuals Achieving contextual ambidexterity requires the organization to have employees with an ambidextrous approach to doing their work. Birkinshaw and Gibson (2004) were some of the first to describe the traits of ambidextrous individuals. They say that ambidextrous employees typically identify and take actions on tasks that are outside their immediate responsibility, build networks to learn and share with others, and are not afraid to initiate ideas on their own behalf (Birkinshaw & Gibson, 2004). Other scholars have followed, arguing that ambidextrous individuals should manage both an entrepreneurial role in searching for new market opportunities and a market-oriented role by having extensive knowledge about customers and competitors (Abebe & Angriawan, 2014). Yet others have focused on learning, claiming that organisations that master a mixture of different learning aspects, namely intra-organisational learning, learning with external partners, and a knowledge-sharing culture, are better equipped to manage ambidexterity (Lin et al., 2013).

Several of the studies on individual ambidexterity have focused on managers and claim that ambidextrous managers must manage many different roles at once (Floyd & Lane, 2000) and must handle conflicting goals (Smith & Tushman, 2005). In addition, they must be skilled at paradoxical thinking (Gibson & Birkinshaw, 2004) and handling seemingly contradictory forces. Paradoxes play an important part in micro-level ambidexterity because exploration and exploitation activities

27

have a largely paradoxical relationship. In addition, and in contrast to the other three models of ambidexterity, contextual ambidexterity cannot resort to using separation as a strategy to handle the paradox.

Separation in time or space are common strategies in handling paradoxes, but another strategy is to accept the paradox and use it to your benefit (Poole & Van de Ven, 1989). Andriopoulos and Lewis (2010) suggest this strategy for organisations seeking to be ambidextrous. In their study of product design companies, they found that it is the ability to embrace and manage paradoxes rather than trying to avoid or resolve them that is the key to ambidexterity. First of all, these organisations manage the paradox of long term and short term survival by having a project portfolio that includes projects targeted for both radical and incremental innovation. Second, these organisations manage the paradox of constraints by shifting between learning about the constraints and client expectations and pulling away from and challenging the constraints. Third, they manage paradoxes by nurturing in their employees the identity of the “practical artist”, that is, people who want to express themselves creatively but still function within the boundaries of projects and deadlines. Finally, they measure team success by looking at the individuals’ performance in order to ensure diversity while at the same time holding individuals responsible for the team’s performance as a means to encourage cohesiveness.

Contexts enabling individual ambidexterity In addition to the personal traits of ambidextrous employees, the ability of these employees to demonstrate their ambidextrous behaviour is dependent on context (Raisch et al., 2009). It has been suggested that a context that uses a mix of “stretch, discipline, support, and trust” is best catered to ambidexterity (Gibson & Birkinshaw, 2004, p.209). Furthermore, an ambidexterity-enabling context should allow and encourage employees to judge for themselves how to allocate time to exploratory and exploitatory activities (Raisch & Birkinshaw, 2008).

Gaps in the ambidexterity literature Many researchers in the field of ambidexterity acknowledge that the micro-level – the study of individuals and teams – is an underdeveloped area and that more research is needed in order to fully understand many of the questions that arise under such circumstances. Some scholars argue that we should focus on what practices need to be put in place in

28

order to manage ambidexterity (Martini et al., 2013), while others hold forward the role of work design (Parker, 2014), how resources are assigned to both exploration and exploitation, or the role of an organisational culture (O'Reilly & Tushman, 2013). In general, studies using empirical data on the individual level are scarce (Good & Michel, 2013). This thesis attempts to contribute to the ambidexterity literature by addressing the individual level of contextually ambidextrous organisations, that is, micro-level contextual ambidexterity, with the aim of exploring challenges related to innovation that these individuals encounter.

2.8 Research questions The research presented in this thesis aims at answering the following questions:

RQ1: What challenges related to innovation are encountered at the micro-level in organizations that have adopted a contextual ambidexterity approach? RQ2: What factors explain the innovation challenges encountered at the micro-level in organizations that have adopted a contextual ambidexterity approach? In addition to the two research questions, the aim of this thesis has been to explore ways to manage the difficulties encountered in order to achieve a fruitful combination of exploration and exploitation. The research setting has been that of R&D departments that have adopted process management in order to make their operations more efficient.

29

3 Research approach and methodology

This chapter provides an overview of the research studies serving as the basis for this thesis. It also provides details of each study to allow the reader to review the quality of data collection and data analysis. Finally, the chapter ends with a discussion about reliability and validity, as well as a reflection on the research journey.

3.1 Overall research design The backbone of this thesis has been three research projects that have been carried out over the course of approximately five years (Table 2). These research projects have been conducted in the context of an industrial PhD project in which the PhD student has been employed in one of the organisations.

The first research project was a survey study that addressed a number of statements regarding the creative climate, project management, and ways of working. This survey was carried out at Scania (referred to as Company A in Table 2) on three occasions and on one occasion in one of the other case organisations. This project served as the data source for Papers A–D.

The second research project was an interview study about the implementation of Lean in R&D and the effects it had on creativity and efficiency. In total, 21 interviewees working in the R&D department of five organisations were interviewed. This project served as the data source for Paper E.

The third research project was an interview study carried out within four design teams at Scania and addressed how employees in a contextually ambidextrous organisation manage to prioritise their time between exploration and exploitation. This project served as the data source for Paper F.

A more thorough description of the studies can be found later in this chapter.

30

Table 2. Overview of the research studies and their set-ups.

Study 1 Study 2 Study 3

Data collection method

Survey (web-based questionnaire) distributed on three occasions

Interviews Interviews

Research focus

Creative climate, ways of working, time pressure, planning, management, etc.

The implementation of Lean in R&D and its effects on operations

The management of exploration and exploitation in a contextually ambidextrous organisation

Units of analysis

Individuals’ views of their team

Individual, team Individual, team, and R&D department

Number of cases

1 case organisations, about 50 design teams

5 case organisations 4 teams in one organisation

Case size About 600 potential respondents on each occasion (range 558–667).

21 interviews. 4-5 interviewees in each organisation.

10 interviews. 1-4 interviewees in each team

Participants Design engineers and other professionals working in R&D

R&D workers and managers

Engineers in R&D

Object Company A (and B*) Company A, C, D, E, F, G

Company A

Papers A, B, C, D E F

*Company B participated in round 2 of the survey study, and that survey had a few additional questions. The questions that were the subject of analysis in the appended

papers were identical in the two companies.

3.1.1 Insider research The research presented in this thesis has been carried out as part of an industrial PhD project. The PhD student conducted research as an employee at Scania while at the same time being affiliated with the KTH Royal Institute of Technology. The PhD student received her salary from the employing company and was expected to continue working within the company after graduation. Other than that, the set-up was similar to that of an academic PhD student and the quality demands on the actual thesis were identical (Kihlander et al., 2011).

31

This type of employment and set-up of the PhD project enabled access to company data that would otherwise be difficult to gain access to. In this particular case, it also enabled continuous informal validation of research results in regular follow-up meetings with a team of senior managers formally tied to the PhD project as a steering committee. The steering committee consisted of four senior representatives from the company and two academic supervisors who met quarterly to discuss different aspects of the research project and to make major decisions. The steering committee meetings, together with regular meetings with the supervisors, helped in analysing the research and its findings in the organisational context. The main part of the scientific supervision, however, was by the academic supervisors.

The special nature of an insider research project also comes with certain challenges that need to be considered. In the initial stage of an insider research project, attention should be given to aligning expectations and demands on the PhD student from its dual actors as well as agreeing on a division of time between research and industry projects (Kihlander et al., 2011). The set-up in this PhD project was that the PhD student spent approximately 80% of her time on research and the remaining 20% on a mix of process-development projects, problem-solving workshops, educating the organisation about innovation, and teaching at the university. What is also important for projects of this type is that there should be a plan for how the research results can be continuously disseminated, primarily within the PhD student’s organization but also to a wider context (Kihlander et al., 2011). In this particular research project, the PhD student fed back knowledge and results to the organisation via both the steering group and seminars at the R&D department of the employing company.

Another aspect to take into consideration is the possible bias that employment within an organisation under study can introduce. In this particular project, the steering committee was responsible for reviewing all publications for information that needed to be kept secret. However, the steering committee never used its right to veto a publication. It is possible that the knowledge about the review process has affected how information about the company has been presented. In practice, however, I have sometimes presented research findings that could be interpreted as unfavourable for Scania and still no comments have been made about this in the review process.

32

The position as an employee with expectations of a future career within the company could also add bias to the research process, and an industrial PhD student needs to reflect on what impact research decisions can have on a career within the organisation. Academic researchers in the field of innovation management, however, experience similar bias. To a large extent, research in the field relies on data from organisations, and depicting a research partner in an unfavourable way could potentially lead to research collaborations not being renewed or even being terminated. In that respect, the bias induced by partnering with organisations that provide data potentially affect both industrial PhD students and PhD students with an employment in academia.

3.1.2 Description of the main case organisation This section describes the context of Scania, the main case organisation, in which the PhD student has been employed. For a description of the other companies in which data for this thesis have been collected, please refer to the appended papers (Paper A and Paper E).

Scania is a global developer and manufacturer in the automotive industry with its main R&D department situated in Sweden. The company employs more than 40,000 people in approximately 100 countries. It is considered an innovative company with several innovation awards in its portfolio such as being named one of the Top 100 Global Innovators by Thompson Reuters (2011). Scania has a wide product development project portfolio, and many of the projects involve hundreds of people working on a single project.

The product development process at Scania is divided into 1) pre-development addressing research, advanced engineering, and concept development, internally called “yellow arrow”, 2) product development that comprises the industrialization process of product development, e.g., concept verification and validation, internally called “green arrow”, and 3) the process of updating products that have already been launched on the market, internally called “red arrow”. The company has chosen not to separate any of these processes into separate organisations, and even on the individual level many employees share their time between pre-development and the industrialization process. This means that the model that the organisation has adopted can be likened to that of a contextually ambidextrous organisation (Birkinshaw & Gibson, 2004). As a means to achieve efficient and effective operations in the R&D

33

department, the case organisation has chosen to adopt a Lean-inspired way of working.

A brief history of the adoption of process-management approaches at Scania The journey of adopting Lean-inspired ways of working in the case organisation began in the production unit and has since then both evolved as a management philosophy and been extended to other parts of the organisation, such as R&D. In the 1990s, the company made efforts to increase things like delivery precision, quality, and cost efficiency in manufacturing using Lean-inspired ways of working. Many employees view this as the starting point of the implementation of Lean. Even before that point in time, however, several initiatives had taken place with the aim of improving operations, including many ideas that were well aligned with Lean thinking. Arriving at the process-management approach that the company holds today has been a journey of gradual adaptations and was not introduced as a “package solution”. The initiative in the 1990s, for example, was followed by further initiatives to complement and build on the changes that had taken place.

In 2003 the first Lean initiative in R&D began. It shared many of the characteristics of the way of working that had been adopted in production, but it also had some features that were specifically developed for R&D such as the inclusion of the principle of product architecture philosophy that is considered important in product development. In 2009, the Lean-inspired process-management model was revised to put further emphasis on creativity, and in 2010 a new wave of implementation of Lean-inspired ways of working was introduced in R&D. This time the process-management model was revised to put increased emphasis on processes orientation rather than a product or technology focus. In 2011, this new version was summarised in a booklet describing all principles and core values. The top management team of R&D also rolled out an educational programme with the goal of teaching the new way of thinking to all employees in R&D.

During the same time, the R&D management team began to focus increasingly on bringing all organisational units into alignment using the booklet describing Lean-inspired ways of working. Regular audits were held to grade design teams on their compliance with the process-management model. Nevertheless, the decentralization of the

34

transformation effort – where each manager trained their own team – and the autonomy of the design teams that culturally has had a strong hold in the organization, led to different ways of implementing the Lean-inspired model. Four years after the 2011 intensification and rejuvenation of the process management model, it can be concluded that while some aspects of the way of working have gained stable ground, other ideas have reached a fairly low level of organisational penetration in R&D.

3.2 Study 1: Survey study The first study was designed as a longitudinal survey study with questionnaires distributed on three occasions over the course of approximately one year. It was planned so that it would follow the progress in the organization as the work with Lean thinking in R&D was intensified and developed in 2011 and the years that followed. The first round of questionnaires served as a baseline, and the two latter rounds were distributed with approximately six months intervals to follow up on the effects of the new ways of working (Figure 7).

Figure 7. An illustration of the questionnaire study laid out over the course of a timeline where the parallel business development effort is also mapped (figure

previously published in Lund, 2012).

3.2.1 Data collection The respondents to the web-based questionnaire received a link in their work email with an accompanying letter describing the study. The questionnaire consisted of three parts. The first addressed background information, the second part consisted of the Creative Climate Questionnaire – CCQ (Ekvall, 1996), and the third part included statements about things like project planning, learning, distribution of time, and goal-setting. Parts one and three of the questionnaire can be found in Appendix A. The second part, the CCQ, is not disclosed for copyright reasons. Each item in the questionnaire was, with a few

35

exceptions, formulated as a statement, and respondents were asked to fill in how well the given statements described their work situation. Statements that were part of the CCQ were answered on a pre-set scale of 0–3 (interval) (Ekvall, 1996), and the statements in part three were answered on a 1–7 Likert-like scale.

The statements were identical in each round of the questionnaire with the exception of the open-ended question and a few statements that were added at the end of the questionnaire in rounds two and three. The response rates of the questionnaire were 83%, 78%, and 76% for each occasion one, two, and three, respectively (Table 3). These levels are considered “very good”, which is the second highest category on a five-category scale ranging from “not acceptable” to “excellent” (Mangione, 1995).

Sampling To obtain a diverse sample of the organization, a question of participation was sent to the senior managers of 13 organizational units. Twelve of these participated with all, or a sample, of the design teams tied to the specific unit. In total, 47 design teams chose to participate in the first round of the study. One team (n = 22) was used as a pilot study to ensure that all statements, including those referring to things like the internal time-reporting system, could be clearly understood. The results from the pilot study were used solely to audit the questionnaire and were not used for further analysis.

During the course of this longitudinal questionnaire, the respondents changed because the sampling was based on teams rather than individuals. The reasons behind this change of respondents were internal and external employee turnover, re-organisations where teams were moved to other units, and re-organisations where teams were split into several smaller teams. A small number of teams was also added along the way, primarily due to an interest in learning more about their creative climate. These added teams were considered to fit the sample group well enough not to alter the sample diversity or the possibilities for extrapolating the results to a larger population. An overview of the questionnaire rounds can be found in Table 3.

36

Table 3. An overview of the three survey rounds (table previously published in Lund, 2012).

Round 1 Round 2 Round 3

Number of respondents 558 667 647

Response rate 83% (460) 78% (519) 76% (489)

Number of participating teams

46 (+1 pilot team) 58 60

Start date 1 April, 2011 15 Nov, 2011 19 April, 2012

End date 3 May, 2011 5 Dec, 2011 9 May, 2012

Number of background questions

12 12 12

Number of closed statements

75 81 83

Number of open questions 0 1 1

Language Swedish Swedish, English Swedish, English

Operationalisation of variables The questionnaire consisted of 75 to 84 statements and questions in each round. For the appended papers included in this thesis, a subset of statements were used for analysis. The information of how each variable in Paper A, B, C, and D has been operationalised can be found in Table 4. Control variables are excluded. In the papers as well as later in this thesis, combinations of these variables are further operationalised as creativity and innovation. In papers addressing idea quantity and idea novelty, the results are, for example, aggregated to revolve around creativity, and papers that address idea implementation and innovation performance are similarly aggregated to revolve around innovation.

37

Table 4. Operationalisation of variables used in paper A-D. Statements marked * are part of the Creative Climate Questionnaire (Ekvall, 1996). The exact phrasing of the

statements has been altered for copyright reasons. Statements marked † were used in the sampling procedure for paper E.

Variable Statement formulation Scale Featured in paper

Idea time There is time available to explore new ideas.* †

0-3 Likert C

Time for learning

In our group, we get to spend time learning new things.

C

Idea quantity Many new ideas are aired here. * 0-3 Likert B, C

Idea novelty Unusual ideas are often aired in discussions.* †

0-3 Likert B, C

Idea implementation

We often realize the ideas that come up. * 0-3 Likert B

Goals for innovation

In our department, explicit goals to be innovative exist.

1-7 Likert B

Clear project goals

Our projects have clearly described goals. 1-7 Likert B

Lack of time I find it very difficult to get time to last for my work assignments

1-7 Likert B, C

Delivery flexibility

It is easy to postpone a delivery. 1-7 Likert C

Obstacles to innovation

What are the main obstacles to innovation in your organisation? Free text A

Innovation performance

In general, our team has been successful in innovation. 1-7 Likert A, B, D

10 dimensions of creative climate

Refer to the CCQ for a full list of statements (Ekvall, 1996) 0-3 Likert D

3.2.2 Data analysis Even though data were collected on three occasions, only cross-sectional analyses were performed on the data. In Paper B and Paper C, correlation analyses and Ordinary Least Square (OLS) linear regression analyses were performed using the statistical analysis programme IBM SPSS statistics 19 (IBM). Paper A used a mix of qualitative analysis of text answers and t-tests that were performed in IBM SPSS statistics 19 (IBM). The analyses in Paper D were performed using Latent Class Analysis (LCA) and covariate analysis in Latent Gold (Vermunt &

38

Magidson, 2005). For a more thorough description of the analyses, please refer to the appended papers.

3.3 Study 2: Interview study on the implementation of Lean in R&D

The second study was carried out as an interview study (multiple case study) with both a within-case and cross-case analysis approach. The interviews had a focus on the implementation of Lean thinking in R&D and how that had affected R&D operations.

3.3.1 Data collection Managers and operational personnel were interviewed from five R&D departments of Swedish companies that had introduced Lean thinking in their R&D operations. In each organisation, 4 or 5 interviews were conducted making 21 interviews in total. The interviews lasted 80–120 minutes each and all but one interview was recorded. The focus of the study was to address how these organisations had managed the implementation of Lean thinking in their R&D operations and how that had affected their performance in terms of issues like creativity, communication, and efficiency. A list of topics addressed in the interviews along with typical questions for each topic can be found in Table 5. Two master’s-level engineering students planned and conducted the interviews as their graduation project (Bergqvist & Helander, 2013) under the supervision of the author of this thesis, who also participated in all interviews in four out of the five organisations.

39

Table 5. A list of topics and typical questions asked in the interviews of study 2. The table has previously been published in Paper E.

Topic Typical questions

Origin and focus of their Lean approach

Who initiated the Lean implementation and why? How has it changed over the years? Who works according to Lean today and in what way do they do it? How has Lean affected work in the R&D organisation?

General description of the R&D work

What project management models are used? How large are the teams? Are teams co-located? What are the challenges in R&D work? Are any specific methods such as set-based concurrent engineering (Sobek, 1999) or modularisation used in R&D?

Improvement work

How do managers promote improvement work? How often are changes made? What kinds of improvements are typically made? How do you control or measure improvement work?

Queues Do you measure queues in your R&D operations? What are your main bottlenecks?

Batch size and rate How frequently are deliveries in the project planned? Do you pay special attention to how large the deliverables are or how much work they require?

Work in progress Do you have a limit for the number of simultaneously active projects? How do you prioritise between projects?

Waste reduction What is your definition of value? How do you work with value and waste? How do you view failure in projects?

Creativity Has the organisation’s creative ability changed since the introduction of Lean? Has your patent count been affected? Do you actively promote creativity or in other ways support it? Do you use methods for creativity? When do people spend time on exploring ideas?

Product development efficiency

What is the largest effect on efficiency that you have gained since Lean was introduced? How has it affected your lead times, product cost, and product quality? How has your resource utilization changed?

3.3.2 Data analysis The interview data from each organisation were first summarised in separate case-reports that were validated with each organisation by sending them to the contact person in each organisation for review. These case-reports were based on a framework for analysis that aimed at identifying each R&D department’s view of the Lean approach and what effects the implementation of Lean in R&D had had on things like creativity, efficiency, and communication. Based on these case-reports, a cross-case analysis was performed where correlations between different themes emerging in the case-reports could be analysed.

40

3.4 Study 3: Prioritisation between exploration and exploitation

The third study was carried out as an interview study at Scania, and the aim was to find diverse examples of how employees in the organisation managed their time between exploratory and exploitatory activities.

3.4.1 Data collection A total of 10 interviews were carried out with members of four different design teams. These design teams were selected because they represented the extremes in the organisation based on two statements in the questionnaire outlined in Study 1, namely how much time that team members experienced spending on ideation and to what extent the team was good at generating unusual ideas. Each interview lasted approximately 60–75 minutes and addressed aspects of time management focusing specifically on managing exploration and exploitation work. Exploration was addressed by asking about aspects of long-term product development, developing for the future, time for creativity, time for exploring opportunities or analysing the technology area, and aspects of pre-development (called “yellow arrow” at Scania). Exploitation, on the other hand, was addressed by asking about delivering in the product-development phase (called “green arrow” at Scania), deliveries in general, and by talking about the “here and now” and “short term” perspectives. “Delivery” is a word used frequently at Scania, and it means a project delivery with a set content and a set date that usually build on existing knowledge and ideas. The interview protocol can be found in Appendix 2. All interviews were conducted by the author of this thesis.

3.4.2 Data analysis The data analysis consisted of transcription of interview recordings using the computer software F4 (Audiotranskription.de, 2005). The passages of the transcriptions that described how the interviewees managed to prioritise between exploratory and exploitatory activities and why they sometimes experienced difficulties doing this were highlighted in the interview protocol. The whole protocol was then sent for validation to each interviewee who was each asked to thoroughly check that the highlighted parts reflected their work situation in a representative way. After the validation procedure, two researchers analysed the data, searching for 1) reasons behind difficulties in prioritising between

41

exploration and exploitation in a way that would enable the generation of innovative output in both the long term and the short term and 2) strategies that the respondents used in order to achieve a fruitful combination of exploration and exploitation.

3.5 Methodological quality assessment Research studies are a product of the methodological choices that guide the research design, data collection, and analysis. This section brings up some important methodological choices and discusses the possible impact those choices may have had on the results. The chapter is divided into the following four quality-assessment categories of research that are outlined in Yin’s book on case-study research (2009): construct validity, internal validity, external validity, and reliability.

3.5.1 Construct validity Construct validity deals with whether the representations of reality, such as variables or compound constructs, capture reality accurately (Yin, 2009). Ensuring construct validity is a challenge in survey-based research, such as the first study presented in this thesis, because construct validity requires both that the questions capture reality well and that respondents understand the questions in the intended way.

Survey research by its nature implies that the researcher assumes that respondents can understand the questions and interpret them in a similar manner (Bryman & Bell, 2007). In order to achieve better construct validity concerning respondents’ understanding of the questions, the questionnaire in Study 1 was tested in a pilot round with 22 respondents working at Scania. These respondents were asked to answer the questionnaire, and after they had completed the questionnaire a group discussion was held about whether they had experienced any difficulties understanding or answering the statements and questions. This type of validation does not capture construct validity in terms of whether the questions managed to capture reality. When it comes to the questions’ representation of reality, this research faces several challenges. The researched concepts of innovation, creativity, and process management are all compound constructs, and several different interpretations exist both in research publications and in “everyday language”. In this thesis, better construct validity is achieved by 1) providing a broad definition of innovation in the questionnaire and 2) to some extent triangulating

42

survey findings with interviews where the constructs could be explained more thoroughly.

Using subjective measures In many cases, variables measured by questionnaires consist of indirect measures of actual behaviour; therefore, the filter of interpretation added by the respondents can possibly skew the results. In study 1 in this thesis, each statement was framed in a way that addressed attitudes and perceptions of behaviour as a way to understand actual behaviour and output. One example in the questionnaire is the statement that addresses idea quantity: “There are many new ideas floating around here.” There is a risk that there is a discrepancy between the perception that there are many ideas in the organisation and the actual number of ideas. Previous research has, however, indicated that subjective measures to a large extent correlate with objective measures of innovation (Alegre et al., 2006). What should also be taken into account is that complex concepts such as the concept of ideas would be difficult to measure truthfully by objective means as well. Ideas can be viewed as a concept formed by mental effort (Newell et al., 1962; Vandenbosch et al., 2006), that is, conceived inside of someone’s mind, and it is difficult to draw the line of what is an idea, a novel idea, a creative idea, or merely a thought. Attempts to measure ideas objectively will likely also suffer from shortcomings. Measuring both idea quantity and idea novelty using subjective measures, as has been done in this thesis, does not ensure construct validity. It is, nevertheless, one feasible way of attempting to capture a concept that is largely a matter of thoughts and thereby difficult to capture.

Lack of time is another construct that is debatable as a subjectively measured construct. I would like to argue that, for the purpose of this thesis, the perceived levels of lack of time are more interesting to assess than objectively trying to measure lack of time, which would involve simply estimating the time a task should take and how much time a person has available. The perceived lack of time and related stress levels will likely affect behaviour more than actual slack levels. In the matter of time available for learning and ideation, I would also like to argue that perception of available time is more interesting than actual time. Actual time available might be scattered or time that is officially envisioned for exploration might in reality be used for other tasks. Asking respondents about their own perception of time for exploration implicitly includes

43

time quality in the variable, such as focused and uninterrupted time, whereas it excludes things like workshops intended for idea generation that were actually spent on more exploitative tasks. The results presented in this thesis are presented along with the assumption that asking people about their attitude and perception of their situation can generate valuable knowledge even though those perceptions might not be “true” in a positivistic view. People’s perceptions of their situation will likely affect their motivation and self-efficacy, which in turn can affect their behaviours and performance.

Assessment of instrument validity Building constructs on single items further complicates matters because such questions can be too general or manage to capture only part of the concept (Bryman & Bell, 2007). This problem is evident in research on extremely complex and compound concepts like innovation and creativity, especially ones that often address processes rather than discrete events. This thesis looks at innovation and creativity, but only occasionally makes use of single items to address those concepts. Instead, this thesis attempts to investigate innovation by addressing many different and more easily described concepts such as time for ideation and learning, the generation of novel ideas, and innovation self-efficacy. This thesis as a whole thus addresses innovation by making use of all the appended papers, whereas each separate survey question only addresses one or a few aspects of innovation. In order to further address construct validity, this research makes use of an established instrument for measuring creative climate (Ekvall, 1996). This instrument, developed within the field of organisational psychology, has been used in other research studies (e.g. Mohamed & Rickards, 1996). It is considered less validated compared to other instruments that measure creative climate (Mathisen & Einarsen, 2004) but is nevertheless frequently used in research. The fact that the CCQ lacks documentation of its psychometric properties (Mathisen & Einarsen, 2004) led to the decision to use the statements in the questionnaire as single items rather than using the whole instrument as advised in publications on the CCQ. Paper D is the only paper where all ten dimensions of the creative climate are used in the analysis. The instrument’s benefits, such as the relevance that the statements had for the research study, were considered to exceed its drawbacks in terms of lack of validation of psychometric properties. The instrument has enabled the measurement of several aspects of creativity such as time for exploration, idea implementation, and idea novelty. One

44

aspect of creativity that the instrument does not address is how valuable the ideas generated are for the organisation.

It should also be acknowledged that the research unit to which this research has been affiliated gained permission to use the CCQ through contact with Ekvall himself. This affected the ease of which the different instruments could be obtained and subsequently affected the choice of instrument.

3.5.2 Internal Validity If construct validity addresses the question of whether reality is truthfully mirrored in the different variables, internal validity regards the question of whether the causal relationships identified between variables reflect actual causal relationships in reality (Yin, 2009). The statistical analyses performed in this research study are cross-sectional, that is, constructs are measured at one point in time. From such data it is difficult to draw causal conclusions other than by using logical reasoning. Therefore, the statistical analyses in this thesis help to identify relationships between exploitatory and exploratory constructs, whereas the qualitative studies, using interviews, help in understanding the causality of relationships. Triangulating both statistical data and interview data has helped to provide a more comprehensive understanding of the phenomena in this thesis.

Another aspect relevant to internal validity is the choice of analysis method. The data collected in Study 1 are mainly of an ordinal type. The analysis used in Paper B and Paper C is OLS linear regression analysis, a method originally developed for interval data. Simulations have, however, shown that OLS linear regression is robust enough to be used with ordinal data (unless the data are exponential in nature) in order to analyse relationships as long as caution is taken in the interpretation of magnitude (Larrabee, 2011).

A common analytical assumption in organisational research that might affect internal validity is the single-handed focus on linear relationships, even though research has shown that several areas are better modelled by curvilinear relationships, such as the relationship between slack and innovation (Nohria & Gulati, 1997). The analyses in this thesis have, when theoretically motivated, tried to control for curvilinear relationships. Such relationships have also been identified (see Paper B). Furthermore, in order to achieve internal validity, the findings in this thesis that are based on data from Scania have been continuously

45

validated with the steering committee formally involved in the PhD project. The steering committee, consisting of four senior managers, has had an active role in both helping the PhD student focus on relevant aspects of R&D work and to validate the findings by comparing them with the situation in their own organisation.

3.5.3 External validity and reliability The question of whether the findings in this thesis could be replicated in another sample, or when repeating the study at a later point in time, is difficult to answer. This thesis attempts to enhance external validity and reliability by limiting the sample and the inference of results to organisations striving for ambidexterity by means of a contextual approach, in other words, achieving both exploration and exploitation in the same organisation. Other organisations with a similar approach that also develop complex products requiring extensive coordination between employees are likely to reach similar results under similar conditions. Organisations are, however, defined by much more than their organisational design and the products they develop. Furthermore, organisations constantly change over time and several change efforts and managerial styles are in place in parallel. Thus, for organisations and complex processes, it is more interesting to study transferability (Lincoln & Guba, 1985) and what aspects of the identified results that can be used in similar contexts. If the knowledge generated in this thesis is valuable for another practical setting, its value outperforms the potential problems with validity and reliability. The implications for a research study in organisations generally not to find simple causal relationships, as in “do A and you will get B”. It is more relevant to view research of this type as “if you are in a similar organisation and situation as the one researched it is worth analysing and monitoring aspect A and aspect B because those are commonly at risk under such circumstances”.

On the topic of external validity and reliability, it is relevant to mention the sampling of cases. First of all, the results described in this thesis are limited in generalizability because the study builds on only a few cases and uses a single case as the main source of data. However, the sampling together with the high response rates enable generalizability to the whole R&D department at Scania. Furthermore, the size of the case organization and the diversity in the teams’ profiles, including things like designing, testing, and pre-development, increase the generalizability to similar types of organisations. The research has been limited to

46

organisations that engage in development of complex products or services that require intensive coordination of a large number of people and where individual employees are expected to work with both exploratory and exploitatory activities in order to achieve a contextually ambidextrous organisation. The findings are also likely generalisable to organisation that operate under similar conditions outside of R&D, but they should be interpreted with care if the above conditions are not fulfilled.

3.6 Reflections on the research process This section provides room for a short personal reflection on the research process. What appears to me to be most important is my role as an employee and PhD student at Scania, the main case of this PhD project. I have spent on average about two days a week at the R&D site working side by side (although not working with) designers, project managers, and business developers in the organisation. To hear about their daily challenges of managing product development’s long-term and short-term goals has given me many clues to why I see the results I see in my research. This makes me an informed researcher, but it also puts me at risk of bias. Some researchers even argue that native researchers are too connected with the research situation and that they cannot be objective enough (Anderson et al., 2007). Yet other researchers say that with the use of reflectivity insider researchers are an important source for a representative picture of organisations (Brannick & Coghlan, 2007). In my own research journey there have been many opportunities for reflection, often with the help of colleagues with insights into the R&D operations. My closest colleagues have told me about their everyday struggles. The research project’s steering committee has helped me reflect on the results from the perspective of several different units within the R&D department. In addition, the many presentations and workshops on the topic of innovation that I have held in the organisation over the years, and the questions asked by the participants at such occasions, have provided me with a picture of the “everyday life” in R&D departments that mirror my results. Given this setting, it would be easy just to describe the everyday struggles of my closest colleagues, but that would give only a small part of the picture. Therefore, it has been very useful to use mixed methods of analysis and to alternate between qualitative and qualitative analyses in order to validate and triangulate my results.

47

To some extent, the role of an insider researcher has provided opportunities for reflections on the difference between insights gained through immersing oneself in the everyday life of an R&D department and knowledge gained through research studies like narrowly framed results obtained by an OLS regression. There are insights that I have gained that would be close to impossible to reach scientifically. I value these highly as clues about how corporations work, but I cannot present them as data in the research studies. One such example is the political side of corporations, which is likely to influence nearly all other organisational factors. Decisions made by top management have sometimes affected factors described in this thesis. The information that I have gained on that topic, often in the form of speculations by colleagues or events described by senior managers with experience from the top executive level, are far from validated enough to serve as input in a research study. They have nevertheless opened my mind to the possibility that decisions that might appear as unwise from many perspectives might actually be sound from the less obvious long-term perspective that includes corporate politics. Corporate politics has not been a topic of this thesis, but knowing that corporate politics might affect organisational behaviour has helped me understand the sometimes odd behaviours and decisions that I have encountered in my research.

3.6.1 Limitations Like all research studies, this research has certain limitations. Some of the most important ones revolve around context. This thesis has studied and contributes to organisations that attempt to reach ambidexterity by achieving exploration and exploitation within one single organisation. It is also limited to organisations that develop products requiring extensive coordination across a large number of individuals involved in multiple projects. This is an important factor because a delay in part of the project could lead to costly delays for the whole project. An organisation that is not under such pressure could likely handle time pressure in the different phases of innovation differently from the ways described in this thesis.

All of the studies that serve as a basis for this thesis were performed in R&D departments in different organisations. This makes the results relevant primarily for R&D departments. The results can, however, be relevant for other organisational units such as marketing, service providers, etc., because the challenge of combining the long and short

48

term is present in a wide array of different organisations, not just those developing technology.

49

4 Summary of appended papers The purpose of this research has been to analyse how R&D employees can manage to successfully combine exploration and exploitation in their everyday work. Six appended papers serve as the backbone of this thesis, and in this chapter the studies and results presented in each paper are summarized.

The appended papers relate to each other in the following way. Paper A analyses innovation in R&D from an overall perspective and addresses obstacles to innovation as experienced by R&D employees. Papers B, C, and D present the results from studies of quantitative questionnaire data and explore the relationships between risk taking and time constraints on the one hand and innovation and creativity on the other . Results from these first four publications served as the foundation of the interview studies that are presented in Paper E and Paper F. Paper F tries to understand and find examples of the patterns that emerged in Papers A–E. Paper F also presents implications of how managers and operational personnel can create time for innovation activities and risk taking in their everyday work. An overview of the papers and the distribution of work between authors can be found in Table 6.

50

Table 6. Distribution of work among the authors of the appended papers.

Paper First Author(s) Co-author(s) Distribution of work between authors

A A. Karlsson K. Lund Stetler

Data were collected in two organisations by one author each. The framework for analysis, the analysis, and authorship were equally distributed between the two authors.

B K. Lund Stetler M. Magnusson Planning of the study was carried out by Lund Stetler under supervision of Magnusson. Data were collected and analysed by Lund Stetler. The article was mainly written by Lund Stetler with Magnusson as a co-author.

C K. Lund Lund Stetler, as a single author, collected the data, performed the analysis, and wrote the paper.

D K. Lund J. Björk M. Magnusson

Planning was carried out by Lund Stetler under supervision of Magnusson. Data were collected by Lund Stetler and analysed by Björk. The article was co-written by the three authors with Lund Stetler as the main contributor.

E M. Helander R. Bergqvist

K. Lund Stetler M. Magnusson

Planning of the study was performed by Helander and Bergqvist under supervision of Lund Stetler and Magnusson. Interviews were conducted by Helander and Bergqvist, and Lund Stetler participated in 16 of the 21 interviews. The conference version of the paper was written by Helander and Bergqvist as first authors and with Lund Stetler and Magnusson as contributing authors. Lund Stetler and Magnusson served as main contributors in developing the conference paper into a journal publication.

F K. Lund R. Glav Planning of the study and interviews were carried out by Lund Stetler. Lund Stetler transcribed the interviews, which were then analysed by Lund Stetler and Glav in collaboration. The paper was co-authored with Lund Stetler as the main contributor.

51

4.1 Paper A Karlsson, A. & Lund Stetler, K. (2013). Frequency versus effect – Obstacles to innovation and their relationship to innovation self-efficacy. The objective of the research presented in Paper A was to explore obstacles to innovation that employees in R&D organisations of two different companies experienced in their everyday work. We were interested in analysing what relationships there were between different types of obstacles and employees’ self-efficacy, in other words, their belief in their own ability to be innovative. A total of 852 respondents received a questionnaire in which they were asked to list the main obstacles to innovation that they experienced in their respective organisations. Their responses were then classified in different categories based on the framework of determinants for innovation as defined by Crossan & Apaydin (2010). The significance of these relationships for the employees’ innovation self-efficacy was then analysed using independent-samples t-tests.

The findings show that the most commonly mentioned obstacles to innovation in both organisations dealt with a lack of time or scarcity of resources. However, the independent-sample t-test comparing the innovation self-efficacy of those who perceived time scarcity as an obstacle to the rest of the population did not show a statistical significance. In contrast, other less frequently identified obstacles showed significant differences in innovation self-efficacy. Significant differences could be seen for the obstacle categories Organizational culture, Goals & strategies, and Project portfolio management. A closer analysis of the responses in the three categories that were significantly related to innovation self-efficacy showed patterns unrelated to the categories. Firstly, a lack of ambition or shared vision was seen as an obstacle to innovation. Not knowing where to focus resources and creative energy made employees feel that their innovative efforts were watered down. Secondly, the respondents described how a lack of risk taking or tendency to play it safe stood in the way of achieving innovative outcomes.

52

4.2 Paper B Lund Stetler, K. & Magnusson, M. (2014). Exploring the tension between clarity and ambiguity in goal setting for innovation. The objective of the study presented in Paper C was to analyse the effects of goal setting in the different phases of innovation. Given that innovation entails phases of exploratory activities as well as phases of exploitatory activities, the study of innovation as a single concept can be misleading. To avoid that problem, we chose instead to study four different aspects of innovation individually. These aspects of innovation were Idea quantity, Idea novelty, Idea implementation, and Innovation performance (self-rated).

The OLS regression analysis showed that a clearly expressed goal of being innovative had a positive relation to all aspects of innovation, but project goal clarity was only positively significantly related to Idea implementation and Innovation performance. The relationship between project goal clarity and Idea novelty showed a curvilinear relationship with a local minimum (Figure 8). This indicated that in order to facilitate the generation of novel ideas, managers should pose either very clear project goals or no project goals (or ambiguous project goals). Lack of time was negatively related to all variables except Idea novelty.

Figure 8. Quadratic curve estimation of the relationship between Project goal clarity

and Idea novelty.

53

This study found an unexpected relationship between Project goal clarity and Idea novelty. When taking into account the indication that Idea novelty also seems rather insensitive to lack of time, the study implies a curious relationship between Idea novelty and aspects of constraints and pressure in product development. These findings suggested the design of the following qualitative interview study, which provided an opportunity to try to understand the ambiguous findings regarding novelty.

4.3 Paper C Lund Stetler, K. (2012). Creativity just in time? The role of delivery precision in product development. The purpose of the study presented in Paper C was to explore the effects that different aspects of time pressure could have on idea generation in a number of design teams in the R&D organization at Scania. The occurrence of many ideas as well as unusual ideas were selected as dependent variables. The aspects of time pressure included the following features: perception of lack of time, possibilities to deviate from planned delivery dates, time available to explore ideas, and time available for learning activities.

The analysis showed that the generation of many ideas was significantly related to time available to explore ideas, but not to any of the other aspects of time pressure. The generation of novel ideas, however, was not significantly related to how much time for idea exploration that people perceived. In contrast, that variable was positively related to a lack of time and negatively related to high possibilities to deviate from planned delivery dates. It should be noted, however, that the significant positive relationship between novel ideas and lack of time only occurred when the analysis controlled for time available to explore ideas. In other words, being hurried need not affect your ability to create novel ideas as long as you can spend your scarce time on idea exploration. Time for learning was also significantly related to the generation of novel ideas. Results from this study should perhaps be interpreted with caution because several of the results did not have high significance levels and the different analysis models did not show identical results. The generation of novel ideas, however, did not seem to be negatively connected to lack of time and strict delivery plans, which had been expected. Instead, a certain level of pressure could potentially be beneficial for the generation of novel ideas.

54

4.4 Paper D Lund, K., Björk, J. & Magnusson, M. (2014). Myopic Creative Climate – The Result of Streamlining in R&D Organizations? In this paper, we set out to explore the creative climate to examine a more fine-grained view of creative climate than the traditionally dichotomous view of either Innovative or Stagnant creative climates. An additional aim was to connect the type of creative climate to Innovation performance on an individual level. The exploratory analysis was performed by means of a Latent Class Analysis (LCA).

The LCA helped identify three different types of creative climate. In addition to the traditional types that show either high or low levels across most dimensions, the analysis revealed a creative climate that showed mixed results across dimensions. We chose to call this third type of creative climate a Myopic Creative Climate due to its low levels of risk taking but high levels of management and co-worker support and engagement amongst employees. This combination likely creates a work environment targeted towards small improvements rather than large innovative steps, that is, a type of short-sighted (myopic) creative climate.

The identification of a third type of creative climate is an important step away from a simplistic dichotomous view of creative climate as either good or poor and indicates that all dimensions of a creative climate are not equally important. In the paper, we also elaborated on whether a Myopic Creative Climate could, perhaps, be beneficial for the later stages of product development. The challenge then becomes to manage the shift between higher and lower levels of certain dimensions of the creative climate, in particular the risk-taking dimension. In an organization employing a contextual ambidexterity approach – as in the case of the studied company – the ability to shift between high and low levels of risk-taking becomes particularly important.

55

4.5 Paper E Bergqvist, R., Helander, M., Lund Stetler, K., & Magnusson, M., Applying Lean in product development – Enabler or inhibitor of creativity?

This study was explorative in nature and aimed at understanding the implementation of Lean in R&D and the effects it might have on R&D performance such as employee creativity and efficiency in development processes. In total, 21 interviews were performed in the R&D organisations of five Swedish companies. The questions addressed aspects of Lean such as value and waste, improvement work, and flow and queues. To relate Lean initiatives to R&D performance, the interviews also assessed whether the interviewees experienced that the implementation of Lean had affected employee creativity, the amount of slack, and process efficiency.

Four out of the five organisations in this study reported that the use of Lean working methods in R&D had improved their communication, and two organisations reported improved quality. The effects of Lean on aspects related to innovation were more difficult to discern. A clear difference could, however, be seen between the firms that had adopted a long-term view of product development and the company that had a more explicit focus on short-term operations. Employees in the organisation that had a short-term view were less happy with the Lean implementation and had also experienced a decline in employee creativity. The other companies did not report a decline in employee creativity. Four of the organisations, however, reported that they had experienced a reduction of both slack and skunk work since the introduction of Lean. Many of the interviewees also acknowledged that skunk work had been a significant source of innovations but that they did not have a plan of how to replace skunk work so that innovation levels could be maintained. This indicates that exploratory activities, such as skunk work, might be unintentionally crowded out when organisations aim at improving control and efficiency in their operations. Because less slack is left in the processes, previously unsanctioned work that led to innovative outcomes will have to be replaced with sanctioned and supported initiatives.

56

4.6 Paper F Lund, K. & Glav, R. (2014) Strategies for managing micro-level contextual ambidexterity – Combining exploration and exploitation in R&D. The main goal of the research presented in this paper was to identify what mechanisms aid or hinder employees as they attempt to make room for innovation activities in an environment of high productivity pressure. Ten semi-structured interviews were conducted with four different teams at Scania. The questions focused on the interviewees’ perceptions of time available for innovation activities and how they manage to make time for these activities despite high demands on productivity.

A number of findings could be elicited from the interviews as being detrimental to the amount of time and effort employees spent on innovation activities. It was found that exploration activities were crowded out by exploitation activities because the latter had higher priority. Employees gave several examples of projects that were late due to a decision early on in the project to deviate from the standard processes put in place to ensure quality and timely completion of the project. In other words, by the time the project started it was known that time plans would be tight and that resources would likely have to be pulled in from other projects. In many cases, these resources were pulled in from less time-sensitive exploration activities without adjusting plans in those projects. Such reallocation of resources led to poor execution of activities early in the product development process that in turn led to expensive and time-consuming rework later in the product-development process.

The interviewees also shared coping strategies they used in order to set aside time for exploration activities despite high pressure from exploitation projects. These coping strategies were in many respects similar to strategies used by ambidextrous firms, with the difference that the interviewees used them on a micro-level. Instead of geographical separation of entire business units (O'Reilly & Tushman, 2004) the employees made space for innovation activities by means of spatial separation, such as sitting in another location in the same building or by shielding themselves from the environment by means of headphones. Instead of temporal separation over long periods of time (Gersick, 1991; Romanelli & Tushman, 1994), the interviewees argued for temporal separation on a daily or weekly basis aided by dedication of time for exploration activities and a strong commitment to such a promise. Other

57

suggestions included removing some resources, often experts or valuable equipment, from the pressure of exploitation activities and instead dedicating those resources to exploration activities.

The individual-level testimonies and the actionable knowledge gained in this study are important contributions to the thesis. They help to make sense of the correlations and significances in the other studies by illustrating everyday challenges facing a person who works in a company employing a contextual ambidexterity model. They also strengthen the notion that processes aimed at achieving exploration in an environment that is mainly targeted towards exploitation would likely benefit from adopting some of the characteristics guiding exploitative activities such as clear goals, strict dedication of resources, and different follow-up measures.

58

59

5 Analysis This chapter provides an analysis of the results from the appended papers and strives to reach a more comprehensive understanding of the results than is possible from the analyses in each paper separately. It starts by describing two challenges of innovation that employees in contextually ambidextrous organisation encounter. These two factors are then described in more depth. To conclude this analysis, the research questions are restated and answers that can be derived from the findings in this research project are summarised.

5.1 Challenges encountered in contextually ambidextrous organisations

The starting point of this thesis has been to analyse what challenges related to innovation are encountered at the micro-level in contextually ambidextrous organisations. Furthermore, the aim has been to analyse why these challenges arise and to analyse ways in which they can be managed. The results presented in the appended papers point mainly to two challenges related to innovation encountered at the micro-level in contextually ambidextrous organisations: time for exploration and room for risk-taking.

5.2 I don’t have time! A key obstacle to innovation? When exploring what employees experienced as the main obstacles to innovation, a lack of time8

8 The correct name for the determinant identified most frequently was “Resource allocation”, but most comments addressed lack of time. For practical reasons, “lack of time” will be used to describe this group of comments in the thesis.

was most commonly mentioned at Scania and Company B (Paper A). The statistical analysis performed on the survey data presented in Paper A pointed to ambiguities in whether a lack of time had an effect on innovation or not, but subsequent studies using both quantitative and qualitative analyses indicated that lack of time for exploratory activities in particular is negatively related to several aspects of innovation (Papers C–F). Examples of exploratory activities in these studies include aspects such as time to explore ideas (Paper C and Paper D), time for pre-development projects (Paper F), and the ability to spend time in so called skunk work, work that aims at coming up with

60

innovations and creative ideas outside of the official project portfolio (Paper E).

Many of the employees who have taken part in these studies by answering surveys or participating as interviewees have testified about the time pressure they are under and how this negatively affects their creative abilities (Paper A, Paper F). This is a common situation in organisations such as the one at Scania, where many projects compete for a common resource pool (Engwall & Jerbrant, 2003; Payne, 1995). The results in the appended papers indicate that such a general lack of time can have a negative correlation with creativity and innovation (Paper B). However, more fine-grained analyses concluded that the relationship is less clear (Paper C). It seems that a lack of time dedicated specifically to exploration activities might affect creativity negatively and that a general lack of time might lead to less time dedicated to these activities (Paper C).

The following section will describe different factors that explain why time for exploration activities is often crowded out in contextually ambidextrous organisations and then describe a number of strategies that individuals use to increase time for exploration.

5.2.1 Reasons for crowding out of exploration activities The reasons behind why organisations are under time pressure can vary widely. In this study, the focus is on the linkage between process management as a means to achieve efficient exploitation and exploration activities. Process management, as well as other organisational factors, help explain why time for exploration is at risk in contextually ambidextrous organisations.

Managing exploration and exploitation on the individual level Process management methods are commonly used to cut lead times and make processes more efficient. But are process-management approaches per se to blame for the lack of time in organisations? The qualitative analysis of the interview transcripts in Paper E did show that employees in the R&D departments in four out of five participating companies experienced less slack in their processes after the introduction of process-management methods that aim to control processes and manage operations more efficiently (Paper E). Interviewees also testified that the room for unsanctioned work, such as skunk work, had been reduced (Paper E).

61

Results in this thesis, however, show signs that while the introduction of process-management frameworks often correlate with an increase in demands of operational effectiveness or cutting back on staff, they are not necessarily the reason exploratory activities are crowded out. Results indicate that it might instead be the pressure to “do more with less” in a more general sense that leads to a crowding out of exploratory work (Paper E, Paper F). Findings from interviews with product developers indicated that, when including the concept of disturbance reduction in the process-management model, employees actually reported an increase in creativity after implementing the Lean ways of working (Paper E). Other interviewees also mentioned that being able to focus on exploratory work was an important factor for innovation and that constant interruptions were one of the main obstacles to exploration in their everyday work (Paper F).

In summary, this indicates that, in order to generate more creative output, there is not necessarily a need to free explorative activities from pressure exerted on the exploratory activities itself, such as deadlines. But the results do suggest that exploratory activities are negatively affected by other types of pressure such as exploitation projects with tight schedules and high priority that “steal” resources from the exploratory activities and by constant interruptions that lead to less focus (Paper E, Paper F). The “stealing” of resources originally planned for exploratory activities is likely an effect of exploration activities being treated as lower priority than exploitation activities (Benner & Tushman, 2003; Levinthal & March, 1993). Signs of this internal hierarchy could be found in sentences like, “It is my experience that pre-development is cut first”, and “The exploitation activities are normally prioritized over pre-development” (Paper F). This hierarchy between activities can cause problems if there are very little slack resources, which is often the case in multi-project settings (Engwall & Jerbrant, 2003).

Unrealistic time plans In Paper F, it was shown that one of the major reasons for the crowding out of exploratory activities at Scania was the unrealistic time plans with which some projects were initiated. Some interviewees stated that they realised at the launch of the project that the plans would not add up if the project were to follow the agreed process. Comments like “the project planning (...) showed that we would miss the targets for both validation vehicles and start of production” indicate that it is partly a planning problem rather than

62

solely the effect of unexpected deviations later in the project (Paper F). This type of naive planning seems to have led to a lack of energy and time left to work on exploratory activities once a large product-development project was up and running. Interviewees stated that “even if people come up with good ideas they don’t always have the energy to implement them”. In addition, employees were often not allowed to spend time on pre-development because the deliverables in the product-development project were behind schedule and had to be prioritised (Paper F).

The pressure to comply with the project schedule also had consequences for the adherence to processes. One interviewee said, “You check what steps you can risk removing to reach a certain deadline, but the process was put in place so that we should have found the problems that we encounter now”. That statement illustrates both the importance of risk-reducing steps in the product-development process that ensure that new technology can be introduced with high quality once it reaches manufacturing and the temptation to skip these steps in order to complete product development projects faster (Wheelwright & Clark, 1992).

Discrepancies in follow-up A third aspect mentioned as a reason behind the crowding out of exploratory activities was the discrepancies in how exploratory and exploitatory activities were followed up. The interviewees repeatedly stated that “there is a sort of hierarchy between things” and that exploitative activities are always prioritised over exploratory activities when in conflict, saying, “Unless you are basically ordered to spend time in pre-development you will put it at lower priority” (Paper F). This view was also aired in replies to the open-ended questions in the survey presented in Paper A. Employees stated, “When you enter the industrialization phase you are under so much pressure to deliver and to progress in the project that there is no space for innovation”. And although some teams that were represented in Paper F had had ambitions to spend more time exploring in their particular field of expertise, such activities had been postponed for years due to the high pace in the later phases of product development where much of the attention was focused on exploitatory activities.

These findings can be compared to the role that autonomy arguably plays in achieving a creative climate in organisations. Autonomy is described as freedom in how to go about realising a certain task, how to make plans, and how to collaborate with others. Such autonomy will arguably lead to more creative outputs (Amabile et al., 1996; Ekvall,

63

1996). Though research studies in the field of creativity stress that autonomy does not necessarily mean that employees get to choose which direction to go, but rather how to achieve a set target (Amabile, 1997), it has previously been shown that a common misconception in industry is that autonomy in relation to innovation is associated with freedom in both direction and ways of working (Bailyn, 1985). Although the research presented in Bailyn’s article is from the 1980s, the results presented in this thesis indicate that industry still often fails to make a distinction between strategic autonomy in terms of what goals to achieve and operational autonomy in terms of how to reach a certain goal (Paper F). Consequently, there is a risk that managers will be reluctant to try to affect innovation positively by using project-management approaches such as setting project goals and ensuring that the project is progressing toward those goals. The results in this study do not suggest any similar reluctance to set clear goals for exploitatory activities. Thus, one underlying reason for why exploration activities are crowded out could be the discrepancy between follow-up of exploitation and follow-up of exploration, and perhaps not the follow-up actions per se. This thesis suggests that in order for exploratory activities to be able to claim their space in the contextually ambidextrous organisation they, too, need to be monitored and followed up.

A combination of factors crowd out exploration The reasons for the crowding out of exploratory activities can thus likely be attributed to the combination of three factors. First, the use of a micro-level contextual ambidexterity model where employees are expected to contribute in both exploration projects and exploitation projects means that exploration and exploitation have to be achieved with a common and shared set of resources (Birkinshaw & Gibson, 2004). An organisation where multiple projects are carried out in parallel often suffers from a lack of resources (Engwall & Jerbrant, 2003). Inequalities in size and urgency in projects further complicates the prioritisation among projects (Payne, 1995). In other words, just by choosing a contextual ambidexterity model an organisation has taken on the challenging task of allocating resources successfully. Second, the observed naive planning in exploitatory projects often results in a situation where employees have more work tasks than can be fit into full-time employment. And third, the hierarchy among different activities assures that, if there is a conflict, the exploitatory activities will be prioritised (Figure 9). Previous research has shown that larger projects

64

are often prioritised over smaller projects and that urgent projects are prioritised over less urgent projects (Payne, 1995). Because projects featuring mostly exploratory activities are often both smaller and less urgent in the short run, it seems reasonable to assume that these project-management mechanisms work in favour of projects featuring exploitatory activities.

Figure 9: A combination of factors explains the crowding out of exploratory activities.

All of these factors work together in creating a situation where exploration is more frequently crowded out. If one of the factors is dealt with, the other two can remain without necessarily causing a lack of exploration. If, for example, the organisation should employ a model of ambidexterity that builds on separation instead of the contextual ambidexterity approach, an individual employee would not be expected to split their time between exploratory and exploitatory activities. And if the planning of projects were made in such a way that the tasks of both exploratory and exploitatory projects could fit into a normal workday, the conflict would also be unlikely to arise. It would still be the case that an employee would have to divide time between the two, but both activities would get the time they needed for managing deliverables. And lastly, if exploration were sometimes to be prioritised over exploitation, it could potentially “hold its ground” in a working environment under constraints. This would, however, lead to other problems because exploitation would then potentially be crowded out.

5.2.2 Point-based knowledge and late deviations An organisation that has impaired its exploratory activities is already at risk in a competitive environment because research has shown that ambidexterity is positively associated with innovation, long-term survival,

65

and other aspects of organisational success (see e.g. Junni et al., 2013; O'Reilly & Tushman, 2013, or the theory chapter of this thesis for a list of relevant literature)9

Sobek, 1999

. There is, however, an additional potential downside for firms that fail to ensure exploratory activities. Spending little time on exploratory activities can lead to a more point-based knowledge approach as opposed to a set-based approach or solution-space approach ( ; Ward et al., 1995). Point-based knowledge is the result of quickly narrowing down to a single solution and trying to realise that solution without exploring alternatives in the solution space. A lack of broad knowledge, often the result of a point-based approach, can have the effect that when problems occur late in the process the organisation has little choice but to start from zero with a new concept or to adapt and change the present concept so that it will meet the performance targets. This is often achieved at a higher cost and a trade-off in other performance factors such as weight or serviceability. Examples of this can be found in the interviews in Paper F where employees stated, “If you start too late you resort to emergency solutions and most often solutions that are expensive and poor for the end customer”.

The emphasis on knowledge creation early on in the product development process can enable a basis for rapid project completion in the later phases (Thomke & Fujimoto, 2000). Spending adequate efforts in learning about the problem and re-using knowledge from earlier projects and delaying decisions until there is a thorough understanding of the proposed solutions have been shown to reduce development time, increase quality (Thomke & Fujimoto, 1999; Ward et al., 1995), and increase productivity in new product development (Cooper & Edgett, 2008). If such knowledge is not in place, late deviations and time-consuming reworks might, in many instances, offset the initial plan for the exploitation project and thereby increase the discrepancy between plan and reality even more. Thus employees who were already behind schedule due to naive initial planning risk falling even more behind schedule because they have to spend time in the later phases of product development creating knowledge and new solutions, or, in other words, actually exploring (Figure 10). While such activities could also count as exploration, they rarely produce the ideas that are keys to future innovations. Some testimonies in the appended papers state that 9 While ambidexterity is generally positively related to aspects of firm performance, a recent a meta-analysis shows that this relationship is moderated by a number of contextual factors (Junni .et al., 2013)

66

creativity that could have been used to invent concepts for future innovations instead went into solving problems that surfaced due to poor planning or poor understanding of the technology. One example of such a statement came from an employee at Scania who said, “We spent very little resources early in the project (...) and now we add resources when we have late deviations when entering production to correct the very problems that we might have found if we would have spent enough on pre-development in the beginning” (Paper F).

Figure 10: An illustration of a self-enforcing combination of factors that leads to

crowding out of exploratory activities.

5.2.3 Making room for exploration The study described in Paper F was designed with hopes, from a research point of view, that a large number of coping strategies for combining exploration and exploitation would be elicited. However, few were identified and many of them related to separation in one way or another. Separation in time or space are merely two of the ways in which paradoxical relationships can be managed. The other ways elicited suggest that we should accept the paradox and deal with the paradoxical concepts in parallel or that we should try to resolve the paradox by, for example, introducing a new concept (Poole & Van de Ven, 1989). These alternative avenues of managing the paradoxical co-existence of two seemingly contradictory concepts within a single organisation are difficult to achieve, and it is not surprising that the suggested coping strategies often build on separation. The following are coping strategies identified in Paper F:

67

Separating oneself from the everyday work environment and all the interruptions that come with it, by for example using headphones to signal being busy or sitting in another location and turning off one’s mobile phone.

Achieving both exploration and exploitation on the unit level, but not the individual level, by full-time dedication of resources, such as experts and test equipment.

Safe-guarding time for both exploration and exploitation by using strict follow up, such as measuring that a certain amount of time is spent on exploratory activities each week.

In addition to these coping strategies, there were a few suggestions about synergy-based coping strategies. For instance, interviewees talked about letting “an idea grow” and how you can use input and lessons learned from the later phases of product development when you initiate new projects.

Changing project goals can undermine creativity In Paper A and Paper B, goal-setting turned up as a promising avenue for making space for exploration. Having an explicitly stated goal that the work unit should produce innovative outcomes was shown to be positively correlated to several aspects of innovation, including the production of many ideas, the production of novel ideas, and idea implementation (Paper B). Furthermore, a clear goal or vision that is stable over time was highlighted as an important factor for innovation. Qualitative analysis of the statements in the group of obstacles coded into the goal-setting category in Paper A indicated that “unclear, insufficient, and constantly changing project definitions” hindered innovation and that there was a need for ambitious goals that point out a clear and innovative direction rather than treating all projects and goals as equally important (Paper A). This is in line with existing research stating that having clear goals often enhances creativity (Amabile, 1998), although some scholars argue that ambiguous goals are a more promising avenue for innovation (Brun et al., 2008). What is more conclusive from previous research is that constantly changing goals will undermine creativity (Amabile, 1998) and lead to lower levels of customer satisfaction and project efficiency (Dvir & Lechler, 2004). The role of goal-setting for innovation as studied in this thesis is further elaborated on later in this chapter.

68

5.3 Dare to win – The role of risk-taking in innovation

Another challenge identified in contextually ambidextrous organisations was risk-taking. In Paper D, an LCA identified three distinct classes of creative climate at the Scania R&D department. Two of these classes followed the model of good creative climate and poor creative climate as suggested by Ekvall (1996) in his description of how to interpret the survey instrument (with the difference that the Conflict dimension was at low levels in the class of poor creative climate at Scania). The third latent class, however, had a different response pattern and it was found that three dimensions, namely Idea time, Risk-taking, and, to some extent, Diversity scored at decreased levels compared to the other dimensions that scored at levels comparable to those with a good creative climate. This latent class also scored significantly lower levels of innovation performance compared to the other latent classes (Paper D). This finding indicates that an appropriate level of risk-taking is an important factor for innovation performance. Furthermore, a lack of risk-taking was identified as an obstacle to innovation in the study presented in Paper A.

The following section of the thesis lays out factors that explain why risk-taking is a challenge in contextually ambidextrous organisations and explores ways of dealing with this challenge.

5.3.1 No novelty, no risk. And no innovation. The generation of innovation entails the realisation of an idea that is novel and useful (Amabile et al., 1996). Introducing novelty into a process almost always comes with a risk of both expected and unexpected problems that call for flexibility and creativity. In other words, the organisation needs to be prepared to take the risks that come with novelty in order to produce innovations (Reinertsen & Shaeffer, 2005), and uncertainty is an inherent part of product development activities that need to be handled rather than avoided (Tatikonda & Rosenthal, 2000). The results in this thesis, however, point to the conclusion that as an organisation focuses on operational effectiveness there is less room for such risks to be taken (Paper A). Aspects of risk-taking were mentioned as a reason for decreased innovation in all of the categories that were significantly related to lower levels of innovation self-efficacy in Paper A. Interviewees in one of the two case companies complained that the project portfolio was deliberately filled with

69

incremental projects, whereas interviewees in the other case company complained that a culture of “lack of courage” and a “fear of taking risks” stood in the way of innovation (Paper A). The LCA in Paper D also showed that low levels of risk-taking were a common characteristic of the creative climate in the studied organisation.

5.3.2 Goal-setting as a means to increase risk-taking The analysis of the open-ended responses in Paper A showed an intimate connection between goal-setting and risk-taking when it comes to factors that influence innovation. The respondents in one case company pointed out low risk-taking in the goal-setting for projects as an important obstacle to innovation, and respondents in the other case company pointed out that the project portfolio was filled with incremental projects (Paper A). These differences can be explained by the fact that the projects at Scania were substantially larger than the projects in Company B. In other words, a goal within a project in the first company can be likened to the goals for an entire project in the other. Nevertheless, these findings reveal the potential role that goal-setting can have for innovation, a relationship that was further explored in Paper B. The findings in Paper B pointed to an unexpected relationship between goal-setting and innovation. The regression analysis showed that a general goal to be innovative was positively related to both ideation and idea implementation, which is in line with existing research (Latham & Locke, 1991; Wood et al., 1987). However, the role of clear project goals was less straightforward. Statistical analysis in Paper B found that clear project goals had a positive and linear relation with the implementation of ideas, such as exploitation activities, but that it had no significant relationship with the generation of many ideas. Furthermore, controlling for curvilinear relationships found that the production of novel ideas had a U-shaped relationship with the existence of clear project goals (Paper B). This indicates that the most novel ideas are produced under circumstances of either clear or ambiguous goals. This finding might resolve some of the contradictions in existing research where two distinct patterns can be discerned: those that say that specific goals might harm performance when exploration is involved (Latham & Locke, 1991) and thus ambiguous goals are preferred for innovation (Brun & Sætre, 2009), and those that claim that specific and ambitious goals are the preferred type for increased innovation performance in many organisations (Sitkin et al., 2011).

70

Promoting different types of novelty The results in Paper B also suggest that ambiguous and clear goals can be used in order to stimulate different types of novelty in ideas. Paper B includes a model of in-depth and wide search that could explain the U-shaped relationship between clear project goals and idea novelty (Figure 11). An in-depth search refers to the process of looking for many different ideas within a small scope, something that would produce more novel ideas in a given direction by digging deeply, and a wide search refers to the process of looking for potential solutions in new and unexpected directions, with the trade-off that the depth of analysis is not always so extensive. In other words, stretch goals (Sitkin et al., 2011) would lead to an in-depth search and ambiguous goals (Brun & Sætre, 2009) would lead to a wide search pattern. These findings are in line with resent research that points to collaboration with external parties being more likely to lead to combinations of new knowledge and information whereas collaboration with close colleagues is more likely to lead to in-depth analysis (Bergendahl & Magnusson, 2015).

Figure 11. A hypothesised model of wide and in-depth search in the generation of ideas

(Paper B).

5.3.3 Delivery precision over innovativeness Finally, the results in this thesis suggest a possible connection between the two factors of exploration time and risk-taking: low levels of slack and demands of strict adherence to project plans can affect how much risk is taken in a project (Singh, 1986). In other words, the results showed that if there was little time to complete the project, which was often the case, the team would choose a more incremental solution over

71

a more ambitious solution (Paper A, Paper F). Interviewees said that “it is difficult to take the time to develop ideas that are related to higher risks. It is easy to pick solutions we already know a lot about in order not to jeopardize the delivery date” (Paper A). Another interviewee stated, “TechX [real name omitted] was eliminated too early. No one had time to work with it so we said that ‘they are too big, there is not enough space to integrate them’. At the same time we knew that the examples of TechX that we had studied were not so large and that they were getting smaller.” This quote indicates that although they knew that the technology was progressing in a direction that could make it feasible they claimed that it would not fit, while the real reason for dropping the concept was lack of time. This way of selecting solutions could potentially be related to the process-management paradigm at Scania in which variability is often viewed as something wasteful rather than a possible source of value creation (Reinertsen, 2009).

The relationship between delivery precision and innovativeness was also explored in Paper C, but the findings in that paper actually suggested that high demands of delivery precision were positively related to idea novelty, although they explained only a small share of the variance in the dependent variable. These findings were controversial. Previous research has shown that strict schedules sometimes make project members take less risk and avoid trying out new things (Ahmed, 1998). A possible explanation for these seemingly contradictory findings is that, when under pressure, teams generate more novel ideas but select more incremental ideas for further development. Another possible explanation that has been brought up in previous research is that people who feel that they are under time pressure sometimes experience that they are more creative, when in fact they are less creative (Amabile et al., 2002).

5.4 Revisiting the research questions The first research question posed in this thesis addressed what challenges related to innovation are encountered at the micro-level in contextually ambidextrous organisations as individuals try to combine exploration and exploitation.

RQ1: What challenges related to innovation are encountered at the micro-level in organisations that have adopted a contextual ambidexterity approach? In answering research question one, we turn to the two challenges of innovation that were identified as particularly influential when aiming for contextual ambidexterity at the micro-level: time for exploration and room for risk-taking. We collect these two exploratory factors under the

72

concept of “micro-level exploration space”. Other factors, such as motivation, support, and autonomy, were not affected or they were affected to a lesser extent in the case organisations in this thesis.

The second research question addressed factors that explain the challenges that arise in contextually ambidextrous organisations.

RQ2: What factors explain the innovation challenges encountered at the micro-level in organisations that have adopted a contextual ambidexterity approach? Several factors were identified in the search for an answer to research question two. It was found that time for exploration was typically crowded out by exploitatory activities because in the later phases of innovation, which are mainly concerned with exploitatory activities, these had a higher priority because project plans had been made in such a way that slack levels were too low to manage both exploratory and exploitatory activities. The higher priority was, in turn, likely an effect of the discrepancies in follow-up measures of exploratory and exploitatory activities. While exploitatory activities, mainly in the later phases of product development, were intensely monitored, no such factors could be identified for exploratory activities in the early phases of innovation.

Furthermore, late deviations in product development projects led to more resources being moved from early to late phases of product development so that the deviations could be handled without moving major deadlines. The research findings also indicated that projects where little time had been spent on exploratory activities in the early phases of product development often had more late deviations than projects where more time had been spent up front.

The other innovation challenge that was identified as potentially problematic in contextually ambidextrous organisations was risk-taking. The findings suggest that employees avoided projects or concepts that were associated with higher novelty levels or higher risk-taking in order not to jeopardize set delivery dates. The strict adherence to delivery dates was in turn a consequence of the Lean process-management approach that had been introduced at Scania.

Furthermore, there was a link between decreased time for exploration activities in the early phases and decreased risk-taking. The results indicated that a failure to explore broadly in the early phases of innovation led to a reluctance to select concepts with high novelty because such a choice would bring with it a higher risk for late

73

deviations. Too little exploration in the early phases also brought with it, according to the interviewees, the problem that finding alternative concepts when concept failure was a fact proved harder and more costly in time, in product cost, and in product quality.

In addition to the two research questions, the aim of this thesis was to explore ways in which the identified challenges can be managed in order to aid a fruitful combination of exploration and exploitation. In short, it can be said that goal-setting is a promising avenue for managing time for exploration and encouraging employees to make room for novel ideas and risk-taking. Furthermore, the research showed that employees often resorted to coping strategies in which they separated themselves from the everyday work environment where exploitation is dominant in order to make space for exploratory activities.

74

75

6 Implications for theory In this chapter, the findings presented in the thesis are compared, contrasted, and discussed in the light of existing research in the field. The implications for theory that can be drawn from this thesis are focused around two areas. First, ideas regarding the pressure exerted on innovation in R&D departments are elaborated on by relating existing research on creativity and operations management to the empirical results presented in this thesis. Second, the implications of separating activities from output when studying ambidexterity are elaborated upon.

6.1 Innovation under pressure The first implication for theory regards how exploratory activities are influenced by different kinds of pressure. Should exploration, for example, be managed by means of deadlines or is autonomy so essential that monitoring progress would harm innovation? When it comes to how exploratory and exploitatory activities should be supported, the recommendations differ from researcher to researcher and are likely also related to the plethora of definitions of ambidexterity. Obstfeld (2012), for example, describes how routine projects and creative projects differ in aspects of planning, management, and decision making, that is, the assumption that routine and creative work should be treated differently. Where routine work is often guided by a set of possible choices as answers to a set of possible situations, non-routine work has less such guidance and answers need to be searched for during the process (Obstfeld, 2012). Management support style is also said to affect exploration and exploitation differently and constantly needs to be adapted to whether the team is working on exploratory or exploitatory tasks (Bledow et al., 2009). To complicate this, some researchers argue that “there is no systematic model that predicts when it is useful to explore and when it is useful to exploit” (Rosing et al., 2011, p.967), which would call for a very responsive leadership style that could potentially lead to conflicting behaviour. Some researchers go so far as to argue that the control systems that are incorporated in order to complete projects in an efficient manner actually stifle innovation (Amabile, 1998; Keegan & Turner, 2002). Other researchers challenge this view. Tatikonda and Rosenthal (2000) argue that all projects, no matter how novel the end-result they aim for, can benefit from project-management

76

approaches such as the use of formal processes and project-management autonomy in terms of adapting to emerging needs in the project and being flexible in assigning resources.

Exploration through increased pressure The results in this thesis suggest that some types of pressure, especially pressure upon exploitatory activities in the R&D operations, tend to lead to less time being spent on exploratory activities. Because exploitatory activities in product development often have priority over exploratory activities, and because deadlines for delivery of exploitatory output are rigid and tight in time, there is a risk that the micro-level exploration space decreases (Paper F). These findings could easily lead to the conclusion that pressure in terms of deadlines and lack of time in R&D will lead to less time spent on exploration and to less exploratory output. What should be pointed out, however, is that the pressure also tends to have a bias towards exploitatory activities in product development. Interviewees even ask for more, not less, follow-up and control of exploratory activities (Paper F) and want a common goal to strive for. With such common goals and a more project management-like approach to exploration, the interviewees argued that the consequences of borrowing resources from exploration would become clearer (Paper F). The reason behind the negative relationship between things like time pressure and exploration seems to be the discrepancy between follow-up procedures for exploration and exploitation rather than absolute pressure levels. The frequent follow-up and tight control in exploitation, supported by methods and work procedures derived from process management, stand in stark contrast to the lack of follow-up of exploratory activities and output found at Scania. A possible remedy for this problem would be to increase the follow-up and pressure on exploration rather than reducing them for exploitation.

Furthermore, interviewees in Paper E stated that both slack and skunk work had gone down after the implementation of the Lean process management approach and that slack and skunk work had previously been important sources of innovative ideas. Previous research has suggested that the reduced slack in organisations is also a natural consequence of new technology such as e-mail, in which information has become instantaneous, with the effect that people also expect immediate responses (Lawson, 2001). They argue that previously naturally occurring slack provided time for reflection and information processing, and that

77

such slack should be deliberately built in to modern organisations (Lawson, 2001). This is a promising suggestion and very much in line with the ideas of this thesis. However, while I agree with the suggestion of building in slack in organisations, I would like to bring this issue one step further. I disagree with the definition of things like reflection and information processing as slack. Slack is generally defined as “resources (...) in excess of the minimum necessary to produce a given level of organizational output” (Nohria & Gulati, 1996, p.1246). I propose that we should increase the status of reflection, discussion, and information processing and that they should be considered necessary ingredients for producing high quality exploratory output rather than as activities that can be performed during “excess time”. Lawson also argues in the final paragraph of the study that we need to “understand that time to think and to fail and to learn is not waste but a critical requirement” (Lawson, 2001, p.133). I suggest that it follows that time spent reflecting and learning should be considered critical exploratory activities rather than slack. This distinction does not change the content of exploratory activities but will likely have an effect on how activities are prioritised when competing for the same resources. As long as the view of exploratory activities as something you do during your excess time is combined with a culture that favours being busy, exploratory activities run a risk of being crowded out by exploitatory activities. This risk is likely to be even larger in an organisation that has adopted process management because slack risks being equated with waste and thus eliminated when processes are made more efficient.

6.1.1 Effective exploration and efficient exploitation Even though exploration and exploitation can likely benefit from the same mechanisms in project management, there are also inherent differences between them. This becomes clear when turning back to the definitions of the concepts. While exploration is related to experimentation, risk-taking, and variation, exploitation is related to efficiency and execution (see e.g. March, 1991 and definitions of exploration and exploitation used in this thesis). Even though both types of activities need resources, equipment, and funding, the operative support of such disparate activities arguably calls for different support mechanisms.

To begin with, the Lean thinking paradigm, which is often used to promote exploitation, states that variability in repetitive processes should

78

be considered waste and that the reduction of variability is desirable (Reinertsen & Shaeffer, 2005). Innovation however, builds on the realisation of novel ideas (Amabile et al., 1996), and novelty is by definition a variation from the previously known. Therefore, innovations cannot be created without accepting a certain level of variability (Reinertsen & Shaeffer, 2005). All variation is, however, not good for exploration either. Variation that comes from poor planning or repeating the same mistakes are not beneficial for exploration (Reinertsen & Shaeffer, 2005), while novel ideas and new discoveries in experiments belong to the desired type of variation. Testing in exploration is also different from testing in exploitation. Exploratory testing should be targeted toward learning and information gathering, whereas validating tests should be targeted towards exploitative success (Reinertsen, 2009).

Different types of pressure for different outcomes Can the good type of exploratory variation be positively affected using pressure mechanisms inspired by the way exploitatory projects are managed? The findings in this thesis suggest that pointing out a clear direction for innovation could help employees in developing innovations (Paper A). Furthermore, a general ambition to be innovative that is articulated within the work unit was found to be positively related to aspects of innovation (Paper B). Goal-setting is an interesting avenue for supporting exploration, but it seems that exploratory goal-setting should not be directly copied from exploitation. Goals of exploitatory activities are often pre-set and often fully defined in terms of things like descriptions of how a product should be assembled in manufacturing. In contrast, goals of exploratory activities are often unknown, or unspecified in nature, such as “learn more about...”. It has been shown that learning goals and “do your best” goals have a more positive influence on innovation (Latham & Locke, 1991; Wood et al., 1987), while specific goals have a more powerful effect on less complex tasks (Locke & Latham, 2006). Some scholars even argue that entirely separate structures in terms of metrics and incentives should be set up for exploration and exploitation (O’Reilly III & Tushman, 2008). The results in this thesis even call attention to differences in goal-setting within different exploratory activities. While idea quantity, for example, was positively related to clear project goals, idea novelty had a U-shaped relationship with the existence of clear project goals (Paper B). This indicates that the intermediate levels of goal clarity are related to the least novel ideas, whereas both unclear or ambiguous goals and clear goals

79

seem to be related to high levels of idea novelty. A possible explanation for this is that ambiguous goals lead to a search for ideas in novel directions, whereas clear project goals lead to a search for novel ideas in the given direction (Paper B). These findings stress that not only do exploration and exploitation differ from each other, but that we likely have to search for different support mechanisms for different aspects within exploration as well.

6.1.2 Pitfalls in applying pressure on exploration What could be the possible side effects of pressuring exploration? In general it can be said that creativity, an important ingredient in exploration, generally suffers under time pressure (Amabile et al., 2002). Only under conditions of high focus, high motivation, and little distraction is creativity likely to be at high levels when under time pressure (Amabile et al., 2002). The results in this thesis point in the same direction as interviewees who have stated that they need focused time in order to be innovative and that constantly being distracted is one of the main threats to innovation (Paper E, Paper F). The results also indicate that time pressure affects concept selection. One interviewee said that product attributes, such as size or cost, are sometimes put forward as the main factor for eliminating a certain concept when in reality the available time (or lack of available time) is the determining factor. The reason behind this behaviour is that lack of time is not considered a valid reason for concept elimination, and that instead of being given more resources, the engineers are asked to work “smarter”. By putting forward reasons like size and price that the organisation accepts, the concept can more easily be eliminated. Not being aware of the effects that time pressure might have on project selection could lead to concepts of inferior quality being selected without the project managers being aware of it.

The findings in this thesis point to aspects of pressure as promising mechanisms, if used with care, in the early phases of product development where exploration activities are ample. The purpose would be to both make sure that early exploratory activities are actually prioritised and that exploratory behaviours, such as experimenting and prototyping, are encouraged. This is in line with some of the innovation paradigms targeted specifically toward the early phases that underline the orientation towards action rather than thinking. Design thinking, for example, stresses the importance of building prototypes and experiments

80

early on and proposes measuring the time that has passed until the first prototype is built as a means to push designers and engineers into experimenting more intensely in the early phases of innovation (Brown, 2008). Research has also shown that, if under time pressure, the solution to a design problem is better if the designers fit several iterations of prototype building into the limited time rather than using all the time to execute one iteration as well as possible (Dow & Klemmer, 2011).

While pressure is an interesting avenue to explore, further research is needed to understand how pressure actually encourages exploratory-promoting behaviour. Furthermore, great care has to be taken to ensure that introducing pressure on both exploitation and exploration still enables working conditions that sustain long-term employee health. It is possible, if not even plausible, that in an organisation where neither exploration nor exploitation is crowded out by the other, something else such as employee well-being or work satisfaction will be crowded out instead.

6.2 Exploitatory activities and exploratory output The second theoretical implication derived from this thesis regards the separation of activities and outputs in the operationalisation of ambidexterity. The operationalisation of ambidexterity as the simultaneous achievement of exploratory and exploitatory output, rather than activities, opens up the door for identifying synergistic effects between exploration and exploitation. Activities are things we do for a duration of time. If we assume that we can classify each activity as either exploratory or exploitatory, the possibilities for simultaneous achievement might be scarce. There are exceptions, such as a factory worker coming up with ideas for improvements of the process while manufacturing products (Victor et al., 2000) or coming up with a new idea (exploration) while delivering a document to another part of the organisation (exploitation), but many of the exploitatory activities we perform require our full attention thereby leaving little cognitive room for exploratory brainwork.

Outputs, in contrast to activities, can come momentarily and do not extend over a period of time. Often, activities are what have led to the output, but there are many examples of exploratory outputs that come as an effect of exploitatory tasks. Examples of such “exploratory exploitation output” include ideas about new concepts that emerge as a result of discussing a current project with colleagues or suppliers,

81

insights about customer needs gained when testing prototypes or products in a product-development project, or general knowledge about things like materials when a specific feature is tested in a product-development project. In addition, the results in this thesis indicate that ideas generated during exploratory activities can be incubated while doing exploitatory work (Paper E). The opposite, where exploratory activities lead to exploitatory output, is also true, although likely to a lesser extent. One such example could be new tests, simulations, or verification methods developed for exploration purposes that can also be used for exploitation.

The studies that serve as the basis of this thesis show examples of exploratory output generated during exploitatory activities. One example mentioned as a means to make room for exploration was to book a creative workshop off site. However, these idea generation workshops were more likely to play the role of idea collection workshops. As one interviewee expressed it, “It really felt as if there was a latent need. People had a lot of ideas waiting to come out.” It turned out that ideas were often conceived during exploitatory activities, but at that specific point in time employees had not been able to act on their ideas. New ideas that came up during the workshop were often derivatives of the ideas that people brought with them to the workshop. Although exploratory output in terms of ideas or new knowledge can be generated during exploitatory activities, time is required to test and further develop an idea. Solely generating exploratory output as a by-product of exploitatory work will likely not be enough to fulfil companies’ innovation needs in a turbulent environment. Exploratory output as a by-product of exploitatory activities might also lead to new ideas and knowledge that follow the same logic as the current customer offerings. In line with existing research, this thesis finds it likely that if an organisation wants to generate exploratory output that is very different from the company’s current offerings, the chances are larger if they can break free from the existing organisational culture and norms and dedicate time to exploring new avenues of technology, business, or the like (O'Reilly & Tushman, 2004). Many R&D departments are, however, required to develop even their more radical innovations in a way that is compatible with existing technology or platforms.

82

6.2.1 Creating and absorbing potential Companies where the next generation of products builds on an existing platform or in other ways follows on the previous generation can particularly benefit from a contextually ambidextrous approach. To illustrate this, I will use the model of knowledge creation and project completion depicted in Figure 12 (adopted from Kennedy et al., 2008). In a traditional view of exploration and exploitation, the exploratory activities create a potential for knowledge and new ideas (yellow arrow) that can be exploited in product development projects (horizontal arrows) (Figure 12, left). I advocate that knowledge can be created during both exploratory activities and exploitatory activities (Figure 12, right). New knowledge can either be fed back to the organisation’s general potential of knowledge or be fed into the next-generation product development project.

Figure 12: The combination of knowledge and idea creation (yellow arrow) and project execution (green arrows) in product development (adopted from Kennedy et al., 2008).

The left figure illustrates the classical separation of exploratory and exploitatory output. The right figure illustrates the synergistic view of exploratory and exploitatory

output presented in this thesis.

In order to make use of output generated in the “wrong” phase of product development or during the “wrong” type of activity, the organisation needs to be able to absorb ideas and knowledge and to make use of them, that is, to possess a type of internal absorptive capacity (Cohen & Levinthal, 1990; Zahra & George, 2002). The synergies of a dual organisation thus lie not only in improving the capability to shift between exploratory and exploitatory activities and to ensure micro-level exploration space, but also to improve the capability to adopt and internalise exploratory output generated during exploitatory activities and exploitatory output generated during exploratory activities.

83

7 Implications for practice This chapter describes how the insights from this research can be interpreted and applied in a practical context. First, possible reasons behind why the micro-level exploratory space is crowded out by exploitatory activities are presented. Thereafter follows a presentation of a number of actions that can be taken to increase the available micro-level exploration space for employees.

7.1 Reasons behind the crowding out of micro-level exploration space

One explanation for the crowding out of the micro-level exploration space can likely be found in the malformation that is created when cherry picking principles in process-management paradigms is combined with aspects of short-termism and seeking quick gains. One example that we find in this thesis is the combination of high delivery precision, which is derived from the principles of Lean thinking (Womack & Jones, 2003), and high resource utilisation levels that are likely a combination of a project portfolio with too many projects and a culture that promotes “being busy”.

To provide an explanation of why this can be problematic, we can use the basic rule of queuing theory that resource utilization is exponentially related to queue length (Kingman, 1966; Little, 1961). Figure 13 graphically shows this relationship. As we can see in that figure, the higher resource utilisation levels we reach the more the queues grow, which in turn stops the flow in the processes. Achieving both high resource utilisation and short queues is nearly impossible.

84

Figure 13: An illustration of the relationship between queue length and resource

utilisation based on Little’s law (Little, 1961) and Kingman’s formula (Kingman, 1966).

The relationship described in Figure 13 is acknowledged in Lean thinking that generally takes on a flow-optimisation perspective such as keeping queues short (Modig & Åhlström, 2012). To achieve that goal, a certain overcapacity or slack is necessary in order to ensure flow. Thus, overpopulating the project portfolio and over-utilising employees should not be considered a Lean way of working. What this thesis revealed, however, is that organisations might try to keep resource utilisation levels high by assigning employees extensive amounts of work while striving for flow efficiency by means of strategies like high demands on delivery precision. In this study, one result of this way of working was that resources were borrowed, or rather “stolen”, from exploratory activities in order to manage the deadlines in the exploitatory activities. This risks leading to low levels of exploratory activities in the early phases of product development where there is more room for exploring novel ideas and new knowledge without necessarily delaying large projects. With that exploratory space removed, the interviewees testified that the organisation often chose concepts and projects with little risk when initiating new projects, which in turn led to cases where novel and potentially valuable ideas were discarded.

7.2 Reclaiming micro-level exploration space This section presents a number of actions that can be taken to promote expansion of the micro-level exploration space. Reclaiming micro-level

85

exploration space entails ensuring enough resources and, perhaps even more importantly, using those resources wisely.

7.2.1 Organisational support structures for exploration Factors on the operative level of an organisation that help to ensure micro-level exploration space are gathered under the concept of organisational support structures. Like the support structure of a building, these factors help to safe-guard exploration from the pressure of other forces, in this case pressure from exploitation activities.

Managing resources and project content differently The results in this thesis suggest that traditional project-management methods, such as specifying the scope of the project and earmarking resources, are also desirable for exploratory activities. Interviewees claimed, for example, that clearer goals that remained intact would lead to better innovation (Paper A) and that a key performance index for a certain amount of resources spent on pre-development would be a promising way to increase micro-level exploration space (Paper F).

Using project-management mechanisms that many managers understand from exploitatory activities in product development would arguably decrease the discrepancy between follow-up measures for exploitatory and exploratory activities. This discrepancy has been identified as a potential reason why exploration activities are crowded out. Decreasing the discrepancy in follow-up measures could aid the process of prioritising between exploratory and exploitatory projects and provide decision-makers with a better view of the long-term consequences of shifting resources from exploration to exploitation.

It follows that we should view exploratory activities such as reflection, lessons learned, experimentation, learning, and discussion as critical ingredients of successful exploratory work rather than activities to be done in “spare time” or when there is slack in the process. If exploratory activities have previously been carried out during unplanned time or as so-called skunk work, organisations should be responsive to the potential decrease in skunk work as an effect of shifts in things like the process-management approach (Paper E). If there is a decrease in skunk work, and skunk work has historically been a source of innovation, the organisation should try to find ways to deliberately make room for exploratory activities that can replace the lost skunk work. It is also important to earmark, not only resources for carrying out exploratory

86

designer work, but also other necessary resources such as test rigs and simulation resources (Paper F).

With that said, there are differences in how exploratory and exploitatory activities should be managed. While exploitatory activities often have both output and a process for achieving that output clearly stated, exploratory activities are often characterised as search behaviour and learning (March, 1991). Maximising learning and maximising productivity are in general driven by different factors. Whereas learning is favoured by a culture of experimentation that accepts failure, maximising productivity is better achieved with the help of things like standardised procedures and clear goals.

In summary, it can be said that

exploratory activities can be planned and followed up in a manner similar to the ways exploitatory activities are planned and followed up. One key is to make sure that resources intended for exploration are actually spent on effective exploration.

exploratory output should be valued differently from exploitatory output. Effective learning and novelty are examples of valuable traits of exploratory output. Efficient execution, throughput time, and verification are valuable traits of exploitatory outputs.

Ensuring quality in the micro-level exploration space Once adequate time for exploration activities is ensured, the task shifts to using that available time efficiently. The findings in this thesis suggest that focused time, fewer distractions, and less switching between tasks are desired qualities for the time spent on exploration (Paper E, Paper F). Research has also shown that switching between projects is related to stress, decreased learning, and poor adherence to schedule (Zika-Viktorsson et al., 2006). One can, therefore, argue that it is important to ensure uninterrupted time for exploration. This does not necessitate weeks of unbooked time in a lab or high levels of slack and absence of deadlines. It can just as well consist of work meetings with a cross-disciplinary team or two engineers elaborating on a concept in front of a computer. It is, however, important that employees have the chance to focus on the task for a longer period of time without having to pause to do exploitatory tasks if good-quality outcomes are to be expected from exploratory work that is under time pressure (Paper E, Paper F, Amabile

87

et al., 2002). One way of facilitating this is to provide employees with an office space specifically assigned and adapted for exploratory tasks. In addition, managers could encourage employees to step wholeheartedly into this space by turning off their phones and avoiding reading emails (Paper F). If these employees are urgently needed in an exploitatory project, the manager knows where to find them.

To summarise, the ideas about ensuring quality in the exploratory space highlight a need for the following:

Enable employees to work on exploration without being interrupted by questions regarding exploitatory activities (Paper E, Paper F). This can be done by providing separate work spaces for exploratory work, by safeguarding employees from being engaged in too many projects at the same time, and by ensuring that key persons in the organisation do not have to take all responsibility for duties like responding to questions from suppliers or the like.

Novelty and risk taking Results presented in this thesis indicate that novelty is affected differently under the pressures of time shortage or deadlines compared to other variables (Paper B, Paper C). Although these results to a certain extent are ambiguous and in need of further research, we can still derive some managerial implications from the findings. First of all, the type of goal-setting that leads to novel ideas seems to be two-fold. Both ambiguous goals and clear goals can lead to higher levels of idea novelty than intermediate levels of goal clarity (Paper B). There are signs that ambiguous or unclear types of goals often lead to a more wide search for solutions and, as a result, tend to produce solutions in novel directions, whereas clear goals tend to lead to a more in-depth search within a limited area, thus being more likely to produce novel solutions in that specific area (Paper B).

Second, the interviewees complained that a portfolio filled with incremental projects, a lack of “brave” projects (Paper A), and a lack of direction for innovation work (Paper A, Paper F) were reasons behind low levels of innovation. A creative climate with low levels of risk-taking and time for ideation was also found to be the most common type of creative climate at Scania, and the innovation levels were rated lower in

88

that group. To some extent, this can be related to the tendency to select incremental ideas for realisation just because those ideas would involve little risk of delaying the project (Paper A, Paper F). Risk of delaying deadlines is unwanted in the later phases of product development, but if this is combined with very low levels of pre-development, as is reported at Scania, the room for experimenting with novel ideas is decreased and, with that, so are the chances of innovation.

Furthermore, as pointed out in the chapter of implications for theory, not all ideas or knowledge are results of exploratory activities. Novel ideas and new knowledge are generated during exploitatory activities as well. To implement these ideas in an on-going exploitatory project could lead to high levels of risk. Therefore, organisations might consider putting mechanisms in place that ensure the effective absorption of such ideas and knowledge into activities where they can be developed and implemented without risking major deadlines or causing late rework.

The following three points summarise the findings related to novelty and risk-taking:

Dare to go for brave ideas during exploration. Foster a culture that can accept high levels of novelty (and thus potential risk) in the early phase of product development where exploratory activities are in focus, while eliminating risk in the later phases of product development that focus mainly on exploitation (Paper A, Paper D, Paper F).

Pose different goals for projects of an exploratory or exploitatory nature, and provide exploratory projects with different aims such as a wide or in-depth search for solutions (Paper B).

Put in place mechanisms that capture exploratory output such as new ideas and new knowledge that might be generated during exploitatory activities, and facilitate the further exploration of these ideas.

89

8 Conclusions Achieving contextual ambidexterity is challenging, and this thesis highlights a number of possible pitfalls along the way. The research in this thesis shows that ensuring sufficient micro-level exploration space like time for exploration and room for novelty and risk-taking is a challenge at the micro-level of contextually ambidextrous organisations. The results indicate that time for exploration likely decreases as a result of three interacting factors: 1) asking employees to cater to both exploration and exploitation, that is, using a contextual ambidexterity approach, 2) scheduling employees at high resource utilisation levels, and 3) prioritising exploitation activities over exploration activities when there is conflict between the two.

The decrease in room for novelty and risk-taking, in turn, is likely to be related to the decrease in time for exploration. The findings suggest that when exploration time is crowded out, the project reaches concept selection with a lack of knowledge about the field of technology. Due to this lack of knowledge, incremental concepts are chosen over more innovative ones so as not to jeopardise the schedule for the project in case unknown problems with the concept occur later in the process.

To increase the micro-level exploration space in organisations, this thesis suggests an increased use of project management methods in order to safeguard time for exploration. This thesis highlights the discrepancy between exploration and exploitation when it comes to follow-up and control levels, rather than the control as such, as one of the main reasons behind the crowding out of the micro-level exploration space. This opens up avenues for studying the use of project-management methods in exploration activities as a means to increase the micro-level exploration space by reducing discrepancies in follow-up levels. This thesis also emphasizes that while the same project-management mechanisms can be used in order to assign resources, different logics need to be applied in achieving results. In other words, exploratory activities should aim at effective exploration rather than efficient exploitation.

This thesis also highlights the need for absorbing ideas and knowledge that are generated during the “wrong” type of activity. There are, for example, knowledge, ideas and concepts that are conceived as a result of

90

exploitatory activities, and this potential for future innovation needs to be absorbed by the organisation so as not to be lost. Absorbing output generated during the “wrong” phase also increases the synergistic effects of a contextual ambidexterity approach.

8.1 Suggestions for future research The main purpose of this thesis has been to analyse challenges of innovation that are encountered at the micro-level in contextually ambidextrous organisations. An additional purpose was to understand the underlying reasons for these challenges. While this thesis makes a few contributions to research in the field, it also points to a number of interesting avenues for future research.

First of all, the relationship between pressure, in terms of deadlines and time shortage, seem to have interesting implications for the creation of novel ideas. This research shows vague and somewhat ambiguous results and also indicates that the creation of novel ideas and the selection of novel ideas are supported by different factors. Therefore, future research should address the following question:

RQ: In what way is novelty in innovation, in terms of idea generation and idea selection, affected by pressure such as deadlines and lack of time?

The second area of interest for future research that can be derived from this thesis is the understanding of how organisational support structures can aid innovation, in particular the micro-level exploration space. This thesis has provided some insights into how goal-setting and follow-up of exploitatory activities affect exploratory activities in a contextually ambidextrous organisation. Results indicate that the discrepancy between follow-up measures for exploratory and exploitatory activities is one reason behind the crowding out of exploratory activities. An interesting field for future research, therefore, would be to study how goal-setting and follow-up of exploratory activities would affect the micro-level exploration space:

RQ: In what way can goal-setting, monitoring, and follow-up measures be used in order to enhance the exploratory space and enable effective exploration in a sustainable way?

91

Third, future research is encouraged to focus on the possible synergistic effects from contextual ambidexterity in terms of exploratory output generated during exploitatory activities and exploitatory output enabled by exploratory activities. This particular area is under-researched and provides a promising way forward for the field of contextual ambidexterity. It is by identifying and making use of synergies that the case for contextual ambidexterity can be made stronger. I suggest that research on this topic should focus both on how the potential of exploratory and exploitatory output generated during the “wrong” activity (e.g. exploratory output generated during an exploitatory activity) can be increased, and how this potential can then be successfully absorbed and turned into value for the organisation.

RQ: How can synergistic effects in terms of output be enhanced in a contextually ambidextrous organisation?

RQ: How can contextually ambidextrous organisations successfully absorb exploratory output generated during exploitatory activities and exploitatory output generated during exploratory activities?

92

93

9 References Abebe, M.A. and Angriawan, A. (2014) Organizational and competitive

influences of exploration and exploitation activities in small firms. Journal of Business Research, 67(3), 339-45.

Adler, P.S., Benner, M., Brunner, D.J., MacDuffie, J.P., Osono, E., Staats, B.R., Takeuchi, H., Tushman, M. and Winter, S.G. (2009) Perspectives on the productivity dilemma. Journal of Operations Management, 27(2), 99-113.

Adler, P.S., Goldoftas, B. and Levine, D.I. (1999) Flexibility versus efficiency? A case study of model changeovers in the Toyota production system. Organization Science, 43-68.

Ahmed, P.K. (1998) Culture and climate for innovation. European Journal of Innovation Management, 1(1), 30-43.

Alegre, J., Lapiedra, R. and Chiva, R. (2006) A measurement scale for product innovation performance. European Journal of Innovation Management, 9(4), 333-46.

Amabile, T. (1998) How to kill creativity. Harvard Business Review, 76(5), 77-87. Amabile, T.M. (1997) Motivating creativity in organizations: On doing what

you love and loving what you do. California Management Review, 40(1), 39-58.

Amabile, T.M., Conti, R., Coon, H., Lazenby, J. and Herron, M. (1996) Assessing the work environment for creativity. Academy of Management Journal, 39(5), 1154-84.

Amabile, T.M., Hadley, C.N. and Kramer, S.J. (2002) Creativity under the gun. Harvard Business Review, 80(8), 52.

Anderson, G.L., Herr, K.G. and Nihlen, A. (2007) Studying your own school: An educator's guide to practitioner action research. Corwin Press, Thousand Oaks, CA.

Andriopoulos, C. and Lewis, M.W. (2010) Managing innovation paradoxes: ambidexterity lessons from leading product design companies. Long Range Planning, 43(1), 104-22.

Audiotranskription.de, F4, Audiotranskription.de Bailyn, L. (1985) Autonomy in the industrial R&D lab. Human Resource

Management, 24(2), 129-46. Benner, M.J. and Tushman, M. (2002) Process management and technological

innovation: A longitudinal study of the photography and paint industries. Administrative Science Quarterly, 47(4), 676-707.

Benner, M.J. and Tushman, M.L. (2003) Exploitation, exploration, and process management: The productivity dilemma revisited. Academy of Management Review, 28(2), 238-56.

94

Bergendahl, M. and Magnusson, M. (2015) Creating ideas for innovation - effects of organizational distance on knowledge creation processes. Creativity and Innovation Management, forthcoming,

Bergqvist, R. and Helander, M. (2013) Applying lean approaches within product development - Enabler or disabler for combining efficiency and creativity?, Master's thesis, KTH Royal Institute of Technology.

Bertels, H., Koen, P. and Kleinschmidt, E. (2008) Lessons from 238 companies: senior management, exploitation, exploration, and corporate entrepreneurship performance. Frontiers of Entrepreneurship Research, 28(19),

Birkinshaw, J. and Gibson, C. (2004) Building ambidexterity into an organization. MIT Sloan Management Review, 45(Summer 2004), 47-55.

Birkinshaw, J. and Gupta, K. (2013) Clarifying the distinctive contribution of ambidexterity to the field of organization studies. The Academy of Management Perspectives, 27(4), 287-98.

Bledow, R., Frese, M., Anderson, N., Erez, M. and Farr, J. (2009) A dialectic perspective on innovation: Conflicting demands, multiple pathways, and ambidexterity. Industrial and Organizational Psychology, 2(3), 305-37.

Blessing, L.T.M. and Chakrabarti, A. (2009) DRM, a design research methodology. Springer, London.

Boer, H. and Gertsen, F. (2003) From continuous improvement to continuous innovation: a (retro)(per) spective. International Journal of Technology Management, 26(8), 805-27.

Bolwijn, P.T. and Kumpe, T. (1990) Manufacturing in the 1990s—productivity, flexibility and innovation. Long Range Planning, 23(4), 44-57.

Brannick, T. and Coghlan, D. (2007) In defense of being “native”: The case for insider academic research. Organizational Research Methods, 10(1), 59-74.

Brown, T. (2008) Design thinking. Harvard Business Review, 86(6), 84. Brun, E. and Sætre, A.S. (2009) Managing ambiguity in new product

development projects. Creativity and Innovation Management, 18(1), 24-34. Brun, E., Sætre, A.S. and Gjelsvik, M. (2008) Benefits of ambiguity in new

product development. International Journal of Innovation and Technology Management, 5(03), 303-19.

Brunner, D., Staats, B., Tushman, M. and Upton, D. (2010) Wellsprings of creation: How perturbation sustains exploration in mature organizations, Harvard Business School Organizational Behavior Unit Working Paper 09-011.

Bryman, A. and Bell, E. (2007) Business research methods. Oxford University Press, New York, NY.

Burns, T. and Stalker, G.M. (1961) The management of innovation. Tavistock Publications, London.

95

Clark, K. and Fujimoto, T. (1991) Product development performance: Strategy, organization, and management in the world auto industry. Harvard Business Press, Cambridge, MA.

Cohen, W.M. and Levinthal, D.A. (1990) Absorptive capacity: a new perspective on learning and innovation. Administrative Science Quarterly, 128-52.

Cooper, R.G. and Edgett, S.J. (2008) Maximizing productivity in product innovation. Research Technology Management, 51(2), 47-58.

Crossan, M.M. and Apaydin, M. (2010) A Multi Dimensional Framework of Organizational Innovation: A Systematic Review of the Literature. Journal of Management Studies, 47(6), 1154-91.

Cusumano, M.A. (1994) The Limits of" Lean". Sloan Management Review, 35, 27-27.

Danneels, E. (2006) Dialogue on the effects of disruptive technology on firms and industries. Journal of Product Innovation Management, 23(1), 2-4.

Davenport, T.H. and Short, J.E. (1990) The new industrial engineering: information technology and business process redesign. Sloan Management Review, 31(4),

Dow, S.P. and Klemmer, S.R. (2011) The efficacy of prototyping under time constraints. In Plattner, H., Meinel, C. and Leifer, L. (eds.), Design thinking. Understand - Improve - Apply, Springer, New York, pp 111-28.

Duncan, R.B. (1976) The ambidextrous organization: Designing dual structures for innovation. The management of organization, 1, 167-88.

Dvir, D. and Lechler, T. (2004) Plans are nothing, changing plans is everything: the impact of changes on project success. Research Policy, 33(1), 1-15.

Ekvall, G. (2000) Management and organizational philosophies and practices as stimulants or blocks to creative behavior: a study of engineers. Creativity and Innovation Management, 9(2), 94-99.

Ekvall, G. (1996) Organizational climate for creativity and innovation. European Journal of Work and Organizational Psychology, 5(1), 105 - 23.

Engwall, M. and Jerbrant, A. (2003) The resource allocation syndrome: the prime challenge of multi-project management? International Journal of Project Management, 21(6), 403-09.

Erlang, A.K. (1909) The theory of probabilities and telephone conversations. Nyt Tidsskrift for Matematik B, 20(33-39), 16.

Fisher, M., Hammond, J., Obermeyer, W. and Raman, A. (1994) Making supply meet demand in an uncertain world. Harvard Business Review May-June, (May-June), 83-93.

Florén, H. and Frishammar, J. (2012) From preliminary ideas to corroborated product definitions: managing the front end of new product development. California Management Review, 54(4), 20-43.

96

Floyd, S.W. and Lane, P.J. (2000) Strategizing throughout the organization: Managing role conflict in strategic renewal. Academy of Management Review, 25(1), 154-77.

Garcia, R. and Calantone, R. (2002) A critical look at technological innovation typology and innovativeness terminology: a literature review. Journal of Product Innovation Management, 19(2), 110-32.

Garcia, R., Calantone, R. and Levine, R. (2003) The role of knowledge in resource allocation to exploration versus exploitation in technologically oriented organizations. Decision Sciences, 34(2), 323-49.

Gersick, C.J. (1991) Revolutionary change theories: A multilevel exploration of the punctuated equilibrium paradigm. Academy of Management Review, 16(1), 10-36.

Gibson, C.B. and Birkinshaw, J. (2004) The antecedents, consequences, and mediating role of organizational ambidexterity. Academy of Management Journal, 47(2), 209-26.

Good, D. and Michel, E.J. (2013) Individual Ambidexterity: exploring and exploiting in dynamic contexts. The Journal of psychology, 147(5), 435-53.

Gupta, A.K., Smith, K.G. and Shalley, C.E. (2006) The interplay between exploration and exploitation. Academy of Management Journal, 49(4), 693-706.

Hackman, J.R. and Wageman, R. (1995) Total quality management: Empirical, conceptual, and practical issues. Administrative Science Quarterly, 309-42.

Hall, J.M. and Johnson, M.E. (2009) When should a process be art, not science? Harvard Business Review, 87(3), 58-65.

Hammer, M. and Champy, J. (1993) Reengineering the corporation: A manifesto for business revolution. Business Horizons, 36(5), 90-91.

Harry, M. and Schroeder, R. (2006) Six sigma: the breakthrough management strategy revolutionizing the world's top corporations. Crown Business, New York, NY.

He, Z.-L. and Wong, P.-K. (2004) Exploration vs. exploitation: An empirical test of the ambidexterity hypothesis. Organization Science, 15(4), 481-94.

Highsmith, J. and Cockburn, A. (2001) Agile software development: the business of innovation. Computer, 34(Copyright 2001, IEE), 120-7.

Holmen, M., Magnusson, M. and McKelvey, M. (2007) What are innovative opportunities? Industry and Innovation, 14(1), 27-45.

IBM, SPSS Statistics 20, International Business Machines Corp. Im, G. and Rai, A. (2008) Knowledge sharing ambidexterity in long-term

interorganizational relationships. Management science, 54(7), 1281-96. Jansen, J., Volberda, H.W. and Van Den Bosch, F. (2005) Exploratory

innovation, exploitative innovation, and ambidexterity: The impact of environmental and organizational antecedents. Schmalenbach Business Review, 57(October 2005), 351-63.

Jansen, J.J., George, G., Van den Bosch, F.A. and Volberda, H.W. (2008) Senior team attributes and organizational ambidexterity: The

97

moderating role of transformational leadership. Journal of Management Studies, 45(5), 982-1007.

Junni, P., Sarala, R., Taras, V. and Tarba, S. (2013) Organizational ambidexterity and performance: A meta-analysis. The Academy of Management Perspectives, 27(4), 299-312.

Kauppila, O.-P. (2010) Creating ambidexterity by integrating and balancing structurally separate interorganizational partnerships. Strategic organization, 8(4), 283-312.

Keegan, A. and Turner, J.R. (2002) The management of innovation in project-based firms. Long Range Planning, 35(4), 367-88.

Kennedy, M., Harmon, K. and Minnock, E. (2008) Ready, set, dominate: implement Toyota's set-based learning for developing products and nobody can catch you. Oaklea Press, Richmond, VA.

Kessler, E.H. and Chakrabarti, A.K. (1996) Innovation speed: a conceptual model of context, antecedents, and outcomes. Academy of Management Review, 21(4), 1143-91.

Kihlander, I., Nilsson, S., Lund, K., Ritzen, S. and Norell Bergendahl, M. (2011) Planning industrial PhD projects in practice: Speaking both "academia" and "practitionese", International Conference on Engineering Design - ICED.

Kingman, J.F.C. (1966) On the algebra of queues. Journal of Applied Probability, 3(2), 285-326.

Krafcik, J.F. (1988) Triumph of the lean production system. Sloan Management Review, 30(1), 41-52.

Lamming, R. (1996) Squaring lean supply with supply chain management. International Journal of Operations & Production Management, 16(2), 183-96.

Larrabee, B.R. (2011) Ordinary least square regression of ordered categorical data: Inferential implications for practice, Master's thesis, Kansas State University.

Latham, G.P. and Locke, E.A. (1991) Self-regulation through goal setting. Organizational behavior and human decision processes, 50(2), 212-47.

Lavie, D. and Rosenkopf, L. (2006) Balancing exploration and exploitation in alliance formation. Academy of Management Journal, 49(4), 797-818.

Lavie, D., Stettner, U. and Tushman, M.L. (2010) Exploration and exploitation within and across organizations. The Academy of Management Annals, 4(1), 109-55.

Lawrence, P.R. and Lorsch, J.W. (1967) Differentiation and integration in complex organizations. Administrative Science Quarterly, 1-47.

Lawson, M.B. (2001) In praise of slack: Time is of the essence. The Academy of Management Executive, 15(3), 125-35.

Leonard-Barton, D. (1992) Core capabilities and core rigidities: A paradox in managing new product development. Strategic management journal, 13(5), 111-25.

98

Levinthal, D.A. and March, J.G. (1993) The myopia of learning. Strategic management journal, 14(S2), 95-112.

Liker, J.K. (2004) The Toyota way: 14 management principles from the world's greatest manufacturer. McGraw-Hill Professional, New York.

Liker, J.K. and Morgan, J.M. (2006) The Toyota way in services: The case of lean product development. Academy of Management Perspectives, 20(2), 5-20.

Lin, H.E., McDonough, E.F., Lin, S.J. and Lin, C.Y.Y. (2013) Managing the exploitation/exploration paradox: The role of a learning capability and innovation ambidexterity. Journal of Product Innovation Management, 30(2), 262-78.

Lincoln, Y. and Guba, E. (1985) Naturalistic inquiry. Sage, London. Little, J.D. (1961) A proof for the queuing formula: L= λ W. Operations research,

9(3), 383-87. Locke, E.A. and Latham, G.P. (2006) New directions in goal-setting theory.

Current Directions in Psychological Science, 15(5), 265-68. Lubatkin, M.H., Simsek, Z., Ling, Y. and Veiga, J.F. (2006) Ambidexterity and

performance in small-to medium-sized firms: The pivotal role of top management team behavioral integration. Journal of management, 32(5), 646-72.

Lund, K. (2012) Process management in R&D - Doom or salvation for creativity? Licentiate thesis, KTH Royal Institute of Technology.

Magnusson, M. and Martini, A. (2008) Dual organisational capabilities: from theory to practice–the next challenge for continuous innovation. International Journal of Technology Management, 42(1), 1-19.

Mangione, T.W. (1995) Mail surveys: Improving the quality. Sage Publications, Incorporated.

March, J.G. (1991) Exploration and exploitation in organizational learning. Organization Science, 71-87.

Martini, A., Laugen, B.T., Gastaldi, L. and Corso, M. (2013) Continuous innovation: towards a paradoxical, ambidextrous combination of exploration and exploitation. International Journal of Technology Management, 61(1), 1-22.

Mathisen, G.E. and Einarsen, S. (2004) A review of instruments assessing creative and innovative environments within organizations. Creativity Research Journal, 16(1), 119-40.

Modig, N. and Åhlström, P. (2012) This is lean. Rheologica Publishing, Halmstad, Sweden.

Mohamed, M.Z. and Rickards, T. (1996) Assessing and comparing the innovativeness and creative climate of firms. Scandinavian Journal of Management, 12(2), 109-21.

Morgan, J.M. and Liker, J.K. (2006) The Toyota product development system : integrating people, process, and technology. Productivity Press, New York, NY.

99

Newell, A., Simon, H.A. and Shaw, J.C. (1962) The process of creative thinking. In Gruber, H.E., Terrell, G. and Wertheimer, M. (eds.), Contemporary Approaches to Creative Thinking, Atherton Press, New York, pp 65–6.

Nobelius, D. (2002) Managing R&D Processes. Focusing on technology development, product development and their interplay. PhD thesis, Chalmers University of Technology.

Nobelius, D. (2004) Towards the sixth generation of R&D management. International Journal of Project Management, 22(5), 369-75.

Nohria, N. and Gulati, R. (1996) Is slack good or bad for innovation? The Academy of Management Journal, 39(5), 1245-64.

Nohria, N. and Gulati, R. (1997) What is the Optimum Amount of Organizational Slack?-A Study of the Relationship between Slack and Innovation in Multinational Firms. European Management Journal, 15(6), 603-11.

Nonaka, I. (1994) A Dynamic Theory of Organizational Knowledge Creation. Organization Science, 5(1), 14-37.

O'Reilly, C. and Tushman, M. (2013) Organizational ambidexterity: Past, present and future. The Academy of Management Perspectives, 24(7), 324-38.

O'Reilly, C.A. and Tushman, M.L. (2004) The ambidextrous organization. Harvard Business Review, 82(4), 74-83.

O'Reilly, C.A. and Tushman, M.L. (2011) Organizational ambidexterity in action: How managers explore and exploit. California Management Review, 53(4), 5-22.

O’Reilly III, C.A. and Tushman, M.L. (2008) Ambidexterity as a dynamic capability: Resolving the innovator's dilemma. Research in organizational behavior, 28, 185-206.

Obstfeld, D. (2012) Creative projects: A less routine approach toward getting new things done. Organization Science, 23(6), 1571-92.

Ohno, T. (1982) Workplace management, 2011 edn. Gemba press, Mukilteo, WA. Parker, S.K. (2014) Beyond motivation: Job and work design for development,

health, ambidexterity, and more. Annual review of psychology, 65, 661-91. Payne, J.H. (1995) Management of multiple simultaneous projects: a state-of-

the-art review. International Journal of Project Management, 13(3), 163-68. Poole, M.S. and Van de Ven, A.H. (1989) Using paradox to build management

and organization theories. Academy of Management Review, 562-78. Raisch, S. and Birkinshaw, J. (2008) Organizational ambidexterity: Antecedents,

outcomes, and moderators. Journal of management, 34(3), 375-409. Raisch, S., Birkinshaw, J., Probst, G. and Tushman, M.L. (2009) Organizational

ambidexterity: Balancing exploitation and exploration for sustained performance. Organization Science, 20(4), 685-95.

Reinertsen, D. and Shaeffer, L. (2005) Making R&D Lean. Research-Technology Management, 48(4), 51-57.

100

Reinertsen, D.G. (2009) The principles of product development flow: second generation lean product development. Celeritas Publishing, Redondo Beach, CA.

Reuters, T. (2011) Top 100 global innovators. Richtnér, A. and Åhlström, P. (2010) Organizational Slack and Knowledge

Creation in Product Development Projects: The Role of Project Deliverables. Creativity and Innovation Management, 19(4), 428-37.

Romanelli, E. and Tushman, M.L. (1994) Organizational transformation as punctuated equilibrium: An empirical test. Academy of Management Journal, 37(5), 1141-66.

Rosing, K., Frese, M. and Bausch, A. (2011) Explaining the heterogeneity of the leadership-innovation relationship: Ambidextrous leadership. The Leadership Quarterly, 22(5), 956-74.

Rothwell, R. (1994) Towards the fifth-generation innovation process. International marketing review, 11(1), 7-31.

Sarkees, M.E. and Hulland, J. (2009) Efficiency and innovation: It is possible to have it all. Business Horizons, 52(1), 45-55.

Schroeder, R.G., Linderman, K., Liedtke, C. and Choo, A.S. (2008) Six Sigma: definition and underlying theory. Journal of Operations Management, 26(4), 536-54.

Shalley, C.E., Zhou, J. and Oldham, G.R. (2004) The effects of personal and contextual characteristics on creativity: where should we go from here? Journal of management, 30(6), 933-58.

Simsek, Z. (2009) Organizational ambidexterity: Towards a multilevel understanding. Journal of Management Studies, 46(4), 597-624.

Singh, J.V. (1986) Performance, slack, and risk taking in organizational decision making. The Academy of Management Journal, 29(3), 562-85.

Sitkin, S.B., See, K.E., Miller, C.C., Lawless, M.W. and Carton, A.M. (2011) The paradox of stretch goals: Organizations in pursuit of the seemingly impossible. Academy of Management Review, 36(3), 544-66.

Sitkin, S.B., Sutcliffe, K.M. and Schroeder, R.G. (1994) Distinguishing control from learning in total quality management: a contingency perspective. Academy of Management Review, 19(3), 537-64.

Smith, W.K. and Tushman, M.L. (2005) Managing strategic contradictions: A top management model for managing innovation streams. Organization Science, 16(5), 522-36.

Sobek, W., Liker (1999) Toyota's Principles of Set-Based Concurrent Engineering. Sloan Management Review, Winter 1999(40, 2), 67-83.

Tatikonda, M.V. and Rosenthal, S.R. (2000) Successful execution of product development projects: Balancing firmness and flexibility in the innovation process. Journal of Operations Management, 18(4), 401-25.

Thomke, S. and Fujimoto, T. (2000) The Effect of “Front‐Loading” Problem‐Solving on Product Development Performance. Journal of Product Innovation Management, 17(2), 128-42.

101

Thomke, S. and Fujimoto, T. (1999) Front-loading problem-solving: implications for development performance and capability, PICMET, IEEE, Portland, pp 234-40.

Turner, N., Swart, J. and Maylor, H. (2013) Mechanisms for managing ambidexterity: a review and research agenda. International Journal of Management Reviews, 15(3), 317-32.

Tushman, M. and O'Reilly, P. (1996) Ambidextrous organizations: Managing evolutionary and revolutionary change. California Management Review, 38(4), 8-30.

Tushman, M.L. and Anderson, P. (1986) Technological discontinuities and organizational environments. Administrative Science Quarterly, 439-65.

Vandenbosch, B., Saatcioglu, A. and Fay, S. (2006) Idea Management: A Systemic View*. Journal of Management Studies, 43(2), 259-88.

Ward, A., Liker, J.K., Cristiano, J.J. and Sobek, D.K. (1995) The Second Toyota Paradox: How Delaying Decisions Can Make Better Cars Faster. Long Range Planning, 28, 129-29.

Waring, J.J. and Bishop, S. (2010) Lean healthcare: Rhetoric, ritual and resistance. Social Science and Medicine, 71(7), 1332-40.

Vermunt, J.K. and Magidson, J., Latent Gold, Statistical Innovations Inc Wheelwright, S.C. (1992) Revolutionizing product development: quantum leaps in speed,

efficiency, and quality. Simon and Schuster, New York, NY. Wheelwright, S.C. and Clark, K.B. (1992) Creating project plans to focus product

development. Harvard Business School Pub., Boston, MA. Victor, B., Boynton, A. and Stephens-Jahng, T. (2000) The effective design of

work under total quality management. Organization Science, 11(1), 102-17. Womack, J.P. and Jones, D.T. (2003) Lean thinking: banish waste and create wealth

in your corporation. Simon and Schuster, London, U.K. Wood, R.E., Mento, A.J. and Locke, E.A. (1987) Task complexity as a

moderator of goal effects: A meta-analysis. Journal of applied Psychology, 72(3), 416-25.

Woodman, R.W., Sawyer, J.E. and Griffin, R.W. (1993) Toward a theory of organizational creativity. Academy of Management Review, 293-321.

Yin, R.K. (2009) Case study research: Design and methods, 4th edn. Sage Publications, Inc., Thousand Oaks, CA.

Zahra, S.A. and George, G. (2002) Absorptive capacity: A review, reconceptualization, and extension. Academy of Management Review, 185-203.

Zika-Viktorsson, A., Sundström, P. and Engwall, M. (2006) Project overload: an exploratory study of work and management in multi-project settings. International Journal of Project Management, 24(5), 385-94.

Appendices Appendix 1. Questionnaire

Appendix 2. Interview guide Paper F

1

Appendix 1. Questionnaire Appendix 1 discloses the questionnaire used in research study 1. Questions with an asterisk (*) were added in the second round of the questionnaire. Questions with two asterisks (**) were added in the third round of the questionnaire. Part two of the questionnaire have not been disclosed from copyright reasons.

Part 1/3 Part 1 consists of a number of questions regarding your background, employment and previous work experience.

State the year you were born

State gender Male Female

State your level of education Secondary education Upper secondary school University 1-3 years University 4-5 years University > 5 years Other ___________

Enter group affiliation If your group is not in the drop-down list, please state group affiliation here:

2

___________

State how many years you have worked at Scania (round up to a full year). If you have worked as a consultant and employee then enter the total time. ___________ State approximately how many years total work experience you have (round up to a full year). Also include such work experience that is not directly relevant to your current position. ___________

State the year you started in your current group (e.g. 2003) ___________

State the year you started in your current position (e.g. 2003) ___________

State the competence level in the technician career to which you belong. (If you belong to different C-levels in different technical areas then state the highest) C1 C2 C3 (Senior) C4 (Expert) Technical Manager Senior Technical Manager Senior Technical Advisor Do not belong to the C-career (applies to e.g. consultants and group managers) Don't know

3

State whether you are an employee at Scania or consultant Employee Consultant Other/Don't know ___________

What type of position do you currently have? (Choose the answer option that you think best corresponds with your position) Designer (hardware or software) Tester (hardware or software) Project manager/Object manager Mechanic Group Manager Other (please specify) ___________

Estimate roughly how you divide your time in Silbertime between yellow, green, red, white and blue time. Yellow ___________ Green ___________ Red ___________ Blue ___________ White ___________

Del 2/3 In part two you shall consider a number of brief statements about the creative climate in your work group. (This part is omitted for copyright reasons.)

4

Part 3/3 Part three consists of a number of questions concerning learning and work procedures. After you have answered the questions you will have the opportunity to give any additional comments.

Innovation is the process that aims to develop new products, services or ways of working. Approximately what percentage of your working hours would you say has been devoted to innovation activities during the past six months? 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

5

Innovation can be categorised in various ways, e.g. a distinction can be made between radical and incremental innovation. Examples of radical innovation are: *The development of completely new products or completely new technology. * Major changes in ways of working that completely replace old ways of working. Examples of incremental innovation are: * Minor improvements to existing products that e.g. lead to improved quality or improved functionality. * Minor changes in ways of working that lead to cost savings, for example. Divide the time you devote to innovation between the four categories below: Radical product innovation ___________ Radical innovation in ways of working ___________ Incremental product innovation ___________ Incremental innovation in ways of working ___________

With regard to the development of products , I have enough time to: Fully

disagree Fully

agree Identifying problem areas

Coming up with new ideas

Implementing ideas

6

With regard to the development of ways of working, I have enough time to: Fully

disagree Fully

agree Identifying problem areas

Coming up with new ideas

Implementing ideas

Fill in how well you think the description corresponds with your work situation. Fully

disagree Fully

agree In our group we get to spend time learning new things.

I make sure I document my lessons. I make sure I share what I learn with my colleagues.

Fill in how well you think the description corresponds with your work situation. Fully

disagree Fully

agree There is scope here to change established ways of working.

I continuously evaluate my work in order to find ways of working that save time and resources.

7

Fill in how well you think the description corresponds with your work situation. Fully

disagree Fully

agree It is easy to postpone a delivery. I find it very difficult to get time to last for my work assignments.

Fill in how well you think the description corresponds with your work situation. Fully

disagree Fully

agree There are clear procedures for how my work shall be performed.

I follow the work procedures that currently exist. We are able develop the work procedures we work to ourselves.

The work procedures that currently exist support me in my work with: Fully

disagree Fully

agree Minor improvements in the work processes

Major changes in the work processes

Minor improvements in the product

Major changes in the product

8

Fill in how well you think the description corresponds with your work situation: Fully

disagree Fully

agree There is scope for a holistic approach when developing new products.

In our group we have being innovative as an explicit objective.

During the past six months we have made efforts aimed at increasing the creativity in our group.

During the past six months we have actively worked with understanding how the principles of R&D Factory relates to the work we do in my work unit.*

During the past six months I have changed the way I work to better comply with the principles of R&D Factory.*

Fill in how well you think the description corresponds with your work situation: Fully

disagree Fully

agree Our projects generally have clearly defined goals.* In general Scania has been successful at innovation.* In general my work unit has been successful at innovation.*

We are successful in implementing new ideas to obtain results in my work unit.*

9

Fill in how well you think the description corresponds with your work situation. Fully

disagree Fully

agree My manager encourages creativity and new ideas ** I have great confidence in the way my manager leads the work unit **

The final question in the three questionnaire rounds were of an open format. Those questions were formulated as below:

First round: In this text box you can add comments about the survey. If you do not wish to give any additional comments, click "Done" below.

Second round: The last question concerns what you experience as the main obstacles for innovation at Scania. Submit your comments in the text box below. In this text box you can also leave comment about the questionnaire as such or post specific questions. If you do not wish to give any additional comments, click "Done".

Third round: The last question is of open character and you submit you answer in the text box below: What makes you innovative in your work? E.g. methods, occasions, other people, challenges, etc. And what do you experience as important in terms of resources, support, encouragement or alike in order for you to be innovative? Preferably list several things. In this text box you can also leave comment about the questionnaire as such or post specific questions. If you do not wish to give any additional comments, click "Done".

Appendix 2. Interview guide Paper F All or a sample of the questions below were asked during the interviews. All interviews started with a short description of the purpose of the study, the role of the interviewer, information about the non-disclosure agreement and information about how the interview data would be handle in terms of validation of quotes with interviewee and keeping interviewees anonymous throughout the process. All interviews ended with an open questions about things the interviewee thought could be relevant but that hadn’t been addressed along with an invitation to contact the interviewer if the interviewee would have any questions in the future.

1. Tell me about your role in the organisation. What are the main tasks in your job? 2. In what way do you work to meet the long term changes in your technology area?

a. Who is responsible for that work? b. How do you know what to develop? Roadmaps? Foresighting? c. Who decides what areas that will become PD projects? What does that process look like?

3. Do you experience that you spend enough time in development of new concepts and new technology to be able to meet the challenges ahead in your technology area?

a. How much time would be required to spend today in order to stand prepared with a competitive offer in 3-10 years? (average per person per week)

b. How much time do you estimate that you spend today? (average per person per week) 4. What do you think are the deciding factors for how much time you spend in projects with a long term

goal compared to short-sighted projects? a. What hinders you from spending an adequate amount of time (if that is the case)? b. How do you do personally find time for idea creation for long term innovation? Tell me about

some tricks that you use to ensure time. 5. Case project: Tell me about the latest project that aimed towards developing something for a 3-10

year horizon. (Recently finalised or close to finalised) a. What was the focus challenge? b. How many were involved? c. When do you plan to finish it or when did it end? d. How come you decided to work with that particular project? e. How did you work with solving the challenge and develop a concept? f. Tell me how you proceeded in selecting the concept that later become the finalised product.

Were there other concepts you discarded? Why in that case? g. Is this representative for how you usually work? h. How did you work with problem solving when challenges occurred during the project after

the final concept had been chosen? 6. Have you ever worked in a project where you felt that you managed to come up with really novel

ideas and managed to realise those? Tell me about the project. What do you think were the determining factors for the success?

a. Did you work with problem solving or ideation in any particular way? b. How did you work with planning, resources, processes, etc?

7. Have you ever worked in a project where you felt that there was too little time for creativity even though the problem would have needed it? Why was that the case? What should you have done instead?

8. Have you ever worked in a project where you prioritised between concepts and you felt that the wrong concept was chosen? Tell me about that project. What should you have done differently? How would that have changed things?

9. Has the way you work with long term product development changed over the last four years? In what way and why?

10. Do you use any process management methods in R&D? (Eg. TQM, lean, six sigma?) Has that affected how you work with product development and innovation? If yes, give examples.